Not Exactly Rocket Science

Blogging on Peer-Reviewed ResearchThe effect that violent films and games have on our minds, and the implications for their place in society, has been a source of much heated debate. Now, a new study looks set to fan the flames even further. Several studies have found that violent media can desensitise people to real acts of violence, but Brad Bushman from the University of Michigan and Craig Anderson from Iowa State University have produced the first evidence that this can actually change a person’s behaviour, affecting their decisions to help others in need.

i-4a2ddf2becbc398bbbd68b0ae651b883-MortalKombat.jpgUsing professional actors, they found that after 20 minutes of playing a violent video game, people who heard a loud fight that ended with an injury took longer to help the victim, and believed that the fight was less serious. Likewise, people who watched a violent film took longer to help an injured woman to pick up her crutches outside the cinema. In the duo’s own words, “Violent media make people numb to the pain and suffering of others.”

In the first study, Bushman and Anderson recruited 320 students under the pretence of studying their tastes in video games. The recruits played 20 minutes of either a violent video game (Carmageddon, Duke Nukem, Mortal Kombat or Future Cop) or a non-violent one (Glider Pro, 3D Pinball, Austin Powers or Tetra Madness).  After playing, the recruits were left alone to fill in a questionnaire. They wrote their favourite type of video game (fighting, fantasy, skill, sports and so on) followed by over 200 questions on such games.

This massive list of questions was a red herring, designed to keep the volunteers busy. Three minutes into the time, Bushman and Anderson played a recording outside the room of two actors fighting about how one stole the other’s partner. The actors’ gender was matched to the recruit inside and the two quickly descended into a shouting match and came to blows.

First actor: You stole her from me. I’m right, and you know it, you loser.

Second actor: Loser? If I’m a loser, why am I dating your ex-girlfriend?

First actor: Okay, that’s it, I don’t have to put up with this shit any longer.

[At this point, the experimenter threw a chair onto the floor and kicked the door to the participant's room.]

Second actor: [groans in pain]

First actor: Ohhhh, did I hurt you?

Second actor: It’s my ankle, you bastard. It’s twisted or something.

First actor: Isn’t that just too bad?

Second actor: I can’t even stand up!

First actor: Don’t look to me for pity.

Second actor: You could at least help me get off the floor.

First actor: You’ve gotta be kidding me. Help you? I’m outta here.

[The first actors slams the door and leaves; the second actor groans in pain for 1.5 minutes]

The first actor had left so there was no physical threat and a  pilot study with 40 students showed that all of them thought that the fight was real. When the recording finished, the experimenter started a stopwatch and measured how long it took the recruit to pop outside to help. Recruits who played violent games were no less likely to help the fictional victim than those playing non-violent ones, but they did take almost five times as long to do so (73 seconds compared to just 16).

Obviously, games differ in ways other than their levels of violence but Bushman and Anderson found that, on average, the players didn’t rate the violent games as any more action-packed, enjoyable, fun, absorbing, arousing, boring, entertaining, exciting, involving, stimulating or addictive than the non-violent ones – they were, obviously, rated as being much more violent.

You could also argue that many of the recruits who played violent games helped eventually, even though they delayed. That’s true but remember also that this is based on the influence of just 20 minutes of gaming, with games randomly assigned to players. Bushman and Anderson also found that just 11% of people whose favourite game genre was “fighting with hands or weapon” helped the actor, while 26% of those who preferred non-violent games did so.

If they didn’t after about 5 minutes, the experimenter stuck his head in and said, “Hi, I’m back. Is everything going alright in here? I just saw someone limping down the hallway. Did something happen here?” If they said they heard the fight, they were asked to rate how serious it was on a scale of 1 to 10, for a formal police report.  The violent gamers were less likely to say that they heard the fight, and those that did rated it as being less serious (by an albeit small but statistically significant degree).

i-96951077d62c0ae08a10b984b3ea0b4c-300.jpgIn their second study, Bushman and Anderson found that in a wholly natural scenario, violent movies delayed helpful behaviour. They staged a scene outside a local cinema, where a young female actor with a cast ankle pretended to drop her crutches and struggled to get them. The duo hid from view and watched how the moviegoers reacted.

They observed 162 people in total and while all of them helped the women, those who had just seen a violent R-rated film took 26% longer to come to her aid than those who had seen a PG-rated film. The gender of the cinema-goers and the size of their groups had no bearing on how long it took them to help. And if the actor dropped her crutches before the showing started, people going into the cinema behaved in the same way regardless of what film they were about to see.

So it’s not that violent movies attract less helpful audience – Bushman and Anderson’s results suggest that the films themselves affect a person’s behaviour. It’s a powerful result – the first “field experiment” showing that violent entertainment can actually affect how long someone takes to help an injured person. To the duo, these studies clearly show that “people exposed to violent media become ‘comfortably numb’ to the pain and suffering of others and are consequently less helpful”.

The studies aren’t flawless by any means. For example, in the cinema experiment, it would have been impossible to properly “blind” the researchers to the trials they were conducting – they must always have known whether the film on show was violent or not, and that could have biased their reactions as they timed the helpful behaviour of the film-goers. The delay in helping was also small (although statistically significant). The same applies to the first experiment’s differences in whether recruits heard the fight or how serious they thought it was.

Even so, both experiments do show that people exposed to violent entertainment take longer to help someone else. Perhaps they feel that violence is more normal or acceptable, or because they feel less sympathy towards victims of violence. Either way the result is the same. It’s also worth bearing in mind that these results are based on short and specific exposures to violent entertainment, which are part of daily life for many people.

The debate about the influence of violent entertainment is an uneasy one, coloured by the influence of billion-dollar industries, the popularity and ubiquity of such media, conservative or liberal values, and whether people enjoy violent games or films in the first place. With such powerful biases at work, getting scientific evidence through studies like this is vitally important – they help to ensure that whatever personal or policy decisions are eventually made, are informed ones.

Reference: Bushman, B., & Anderson, C. (2009). Comfortably Numb: Desensitizing Effects of Violent Media on Helping Others Psychological Science, 20 (3), 273-277 DOI: 10.1111/j.1467-9280.2009.02287.x

Subscribe to the feed

Comments

  1. #1 jay
    March 16, 2009

    Unfortunately mentioning “billion dollar industries” etc, you fail to mention the single most important issue: freedom of expression and freedom of people to choose what they want to watch.

    We don’t need government policies, whether ‘informed’ by science or not deciding that censorship is ok because it’s for the ‘greater good’

  2. #2 John
    March 16, 2009

    Having not read the study, I wonder why they used such old games (I played carmageddon as a freshman in 1997).

  3. #3 MattK
    March 16, 2009

    This is extremely interesting. The most surprising thing to me was that there was no detected difference between audiences before the films. Assuming that people who watch a particular violent movie also watch other violent movies this suggests that the effect is not permanent (they had recovered until receiving their second fix).

    jay, who said anything about government censorship?

  4. #4 valdemar
    March 16, 2009

    Okay, Jay, I choose to watch the following:

    1. A full five seasons of Firefly with loads of special guest stars and at least one really funny orgy scene.

    2. A tear-jerking but heartwarming animation about a three-legged squirrel that becomes professor of hermeneutics at Leipzig.

    3. A documentary series about remarkable sandwiches down the centuries, hosted by Meatloaf and Helena Bonham-Carter.

    What? None of them available? Clearly this ‘freedom of choice’ thing needs a bit more work. It seems our freedom to choose what we ‘want to watch’ is limited by all sorts of factors beyond the discerning individual’s control!

  5. #5 Lilian Nattel
    March 16, 2009

    I think this is an important study. There’s long been a debate about the impact of violence in entertainment and research like this is really important to answer that question. While people are free to choose their entertainment, they may not realize what kind of effects they’re choosing in doing so and they have the right to that information. The entertainment industry is interested in making money. They will just as happily sell non-violent games and put more effort into making them exciting if that is what people will buy.

  6. #6 jay
    March 16, 2009

    jay, who said anything about government censorship?

    Well that seems to be an issue here:

    ensure that whatever personal or policy decisions are eventually made, are informed ones.

    Also from another comment

    Clearly this ‘freedom of choice’ thing needs a bit more work. It seems our freedom to choose what we ‘want to watch’ is limited by all sorts of factors beyond the discerning individual’s control!

    That is a bit of a straw man argument. There is a difference between something not having been produced because someone didn’t bother to produce it, and the issue addressed by our Constitution where the expression is suppressed by government action. And that is specifically my area of concern, because this type of thing is often used as an attempted justification for censorship. All attempts at censorship are always ‘for the good of the reader or for the good or society’. This has been the claim by the church and governments all the way back

    There is no need for a constitutional amendment to protect speech that no one finds offensive. It’s the stuff that someone doesn’t like that requires protection

  7. #7 Erasmussimo
    March 16, 2009

    I have some expertise in the video games field, and I believe that the violence issue is not as important as the change in state of consciousness induced by videogames, violent or nonviolent. I believe that there are substantial similarities between the state of consciousness of a gamer who is “in the zone” and that of a person under the influence of some mind-altering drugs.

  8. #8 Keivn H
    March 16, 2009

    The study also doesn’t deal with how long this effect lasts, or if it builds with multiple exposures, but like you said a very interesting article none the less.

  9. #9 jay
    March 16, 2009

    I should comment also that I personally do not like violent films. I don’t watch war films, horror films, crime films, I have walked out of the theater when films turned unexpectedly violent. But that’s a choice that I made. That’s what personal choice and personal responsibility is all about

    I guess I am reacting strongly and pro-actively to the implicit suggestion that freedoms are somehow contingent on whether we can demonstrate an ‘overall good’ or not, a sort of collectivist measure. Unfortunately even among the scientific community there are quite a few who seem to accept looking at an overall statistic as more important than personal freedom and responsibility.

  10. #10 MattK
    March 16, 2009

    So jay, are you suggesting that sociologists and psychologists should not try to learn anything about culture, society, or behaviour because learning about something is just a prelude to taking away all of your freedoms? Is the concept of personal freedoms so fragile that it will crumble as soon as we are not totally ignorant about the influence of violent movies? I think that you are reading way more into this than what is there. Nowhere does it advocate censorship. Nowhere does it advocate for abandoning the concept of personal responsibility. Where are you getting this stuff?

  11. #11 ian
    March 16, 2009

    maybe the MPAA should read this test. they’re more likely to allow violence in films if the gory consequences are played down or not shown. did any of the movies in this study show any link between violence and human suffering? if so, did that affect how desensitized the viewers were?
    personally i don’t like the amount of power the MPAA has over which movies get shown in theaters. i’m not opposed to having a ratings system as merely a tool to help parents, but it might be better if the ratings took these studies into account rather than reflecting the personal opinions of a handful of board members.

  12. #12 Foos84
    March 16, 2009

    This study is compelling. However, I would be careful in saying that the violent media ‘numbs’ people to the suffering of others just because they are less likely to help. Less helping behavior doesn’t necessarily represent a lack of perception of suffering, just a different response to it.

    I’d really like to see if exposure to violent media makes a difference in the activation of mirror neurons. Then we could say with more certainty how ‘numb’ these subjects were.

  13. #13 Nicholas FitzGerald
    March 16, 2009

    “Glider Pro, 3D Pinball, Austin Powers or Tetra Madness”

    Those games are all really lame… maybe the “violent” game group took longer to respond because they were having more fun?

  14. #14 Gotchaye
    March 16, 2009

    I’m mostly curious as to whether the 11% and 26% help rates for the violent and non-violent preferers was controlled for other variables. You’ve got to figure that that correlates really strongly with gender, among other things. This is especially relevant seeing as how the actors were the same sex as the subject. I wouldn’t be at all surprised if you saw something very similar in help rates for males helping other males that they believe to be injured and females helping other females that they believe to be injured.

    The movie thing does seem to suggest that the effect is temporary.

  15. #15 Azkyroth
    March 16, 2009

    So jay, are you suggesting that sociologists and psychologists should not try to learn anything about culture, society, or behaviour because learning about something is just a prelude to taking away all of your freedoms? Is the concept of personal freedoms so fragile that it will crumble as soon as we are not totally ignorant about the influence of violent movies? I think that you are reading way more into this than what is there. Nowhere does it advocate censorship. Nowhere does it advocate for abandoning the concept of personal responsibility. Where are you getting this stuff?

    Jay’s response represents a technically unreasonable but understandable reflex that has been required in response to numerous examples of people who want to believe that exposure to media violence is categorically pernicious taking this sort of study and running with it, pretending it shows far more than it does far more conclusively in order to justify ill-considered and ultimately destructive social movements or legal restrictions. Think along the lines of a combat veteran who’s been shot at enough that he freaks out and dives into a defensive position, grabbing for guns that (hopefully) aren’t there, after a car backfires nearby.

    Personally, I’ve been enjoying violent entertainment for quite some time and still seem to be more inclined to help people than most of the people around me. And I remain suspicious of studies that purport to prove what large numbers of people desperately want to believe.

    Also, while I don’t feel like I know enough about sociological and psychological research to evaluate this kind of experiment design, I’m wondering if someone with some relevant expertise could detail how this research stacks up against the criticisms of the general body of research on “media effects” outlined in this article.

  16. #16 Jane Blogs
    March 17, 2009

    Would there be a difference between fictional violence i.e. computer games or films and real violence as portrayed in t.v. news coverage?

  17. #17 toro
    March 17, 2009

    It seems obvious to me that someone’s appreciation of a specific situation depends on the current context and/or the one encountered shortly before. How long does the effect last is the real question.

    Also I cannot believe the chosen non-violent games can be rated as action-packed and emotionally-engaging as the violent ones (..except tetra-madness maybe).
    Some racing or sports game where you have the adrenaline without the violence would have been better choices.

  18. #18 idlemind
    March 17, 2009

    From the Phil Ochs song “A Small Circle of Friends”:

    Look outside the window, there’s a woman being grabbed
    They’ve dragged her to the bushes and now she’s being stabbed
    Maybe we should call the cops and try to stop the pain
    But Monopoly is so much fun, I’d hate to blow the game

  19. #19 Neuromancer
    March 17, 2009

    I would tend to agree withe the precedent commentator who said that the non-violent games don’t seem very interesting and immersive, but the study says the opposite (see ยง6).

    The problem I see is that the study shows that these violent videogames have an impact on the behavior of players in the described situation, but provides no argument to say that the effect is linked to the specific fact that the task is about helping someone. Maybe these violent, immersive videogames have a general effect on players that make them react slower to real-life social situations after playing, but there is no proof it has something to do with empathy or with helping people specifically. We should compare with another situation which asks for a quick response but has nothing to do with helping people.

  20. #20 CJM
    March 17, 2009

    (Forgive me if I’m repeating another commenter, I didn’t read them all.)

    There have been many similar controlled psychological studies done about violent/objectifying pornography, done on both men and women, regular viewers and not. The short term results about attitudes toward women showed similar differences as you describe about violent video games.

    What you fail to mention is any study done long-term about these gamers’ attitudes. The violent pornography studies that conducted follow-ups found that the attitude shifts were short-lived (between 20 minutes to 2 hours, depending on the study), and the test and control groups faded back to having identical attitudes.

    I’d like to know if this study you mention did any sort of long-term follow up like the ones I’m talking about. I’d be helpful and cite them, but I don’t have my psych textbook with me at the moment.

    Personally, it’s hard for me to believe that playing violent video games has such a substantial long-term effect on people. I have many friends who play these types of games, and they are some of the nicest, generous and non-violent people I know. Obviously I do not have the resources to do a controlled study, but I am skeptical based on these pornography studies too.

    And I also know of other studies that have been done showing that kids who play video games in general, even violent ones, are less likely to act out violently toward people in real life. There is some speculation that the games provide an outlet for already-present aggression that those kids might otherwise choose to express toward real people. (Again, no cite, sorry.)

    So again, I’d like to see a longer-term study of these students’ attitudes toward violence.

  21. #21 Rr
    March 17, 2009

    Hmm… I’d like to believe I’d run to the aid of anyone who needs help, as long as they’re in a situation that isn’t threatening for me, regardless of what I’ve just been watching. But then again I suspect that this sort of thing weighs heavier on my mind than the average person. I wonder what the differences in behavior would be between people from different backgrounds, e.g. country, class, urban/rural, war zones/ peace zones, abused/normal, and so on, for violent media.

  22. #22 grrlpup
    March 17, 2009

    Is anyone else surprised this was okayed by the Human Subjects committee? I’m not in the field, but I know I would be stressed out if this scenario happened right outside the room I was in.

  23. #23 Alex Deam
    March 17, 2009

    This is an interesting study, but really there shouldn’t be any policy implications from it or any similar studies with similar results. I could at least understand a cry for censorship (while probably not supporting it) if it was shown that violent media make people more violent, but this study isn’t doing that. It’s showing that violent media makes people act less like good samaritan. Sure, it’s a bad thing not to go help someone, and I agree such inaction is morally, but why should it be illegal? Why should people be forced to help other people?

  24. #24 Tony Jeremiah
    March 18, 2009

    they found that after 20 minutes of playing a violent video game, people who heard a loud fight that ended with an injury took longer to help the victim, and believed that the fight was less serious. Likewise, people who watched a violent film took longer to help an injured woman to pick up her crutches outside the cinema. In the duo’s own words, “Violent media make people numb to the pain and suffering of others.

    This sounds alot like the bystander effect, with the only difference being its inducement by watching violent video games. The effect has been explained by Latane and Darley (1970) as a five step decision-making process for helping in emergencies (noticing an emergency, interpreting it as an emergency, taking responsibility to help, deciding to help, actually helping). It sounds like watching violent video games short-circuits stage 2 (interpreting an emergency as an emergency), or, puts the brain in a ‘loop’/shock and trauma state that prevents and/or slows helping according to law enforcement officials on the website. This is supported by their results showing, ‘…violent gamers were less likely to say that they heard the fight, and those that did rated it as being less serious…’.

    The authors’ interpretation that violent video games makes people numb to the suffering of others is not directly supported by the data presented here because no questions were asked about their emotional state while witnessing the accident (e.g., what were their feelings for the persons while witnessing the incident). A more neutral intepretation of their data seems to be the same as the one used to explain the bystander effect; that is,a slowing of decision-making processes associated with helping.

    Bushman and Anderson also found that just 11% of people whose favourite game genre was “fighting with hands or weapon” helped the actor, while 26% of those who preferred non-violent games did so.

    If I’m understanding this right, this seems to undermine a media effect somewhat. This data suggests a trait effect such that persons who report liking video games with specific violent content are less likely to have helped than persons who reported not liking violent video games.

    What would have been interesting is to examine the interaction of both trait and media effects. For example, was there a statistical difference in helping for study participants who saw the violent film in the context of those who reported liking violent video games vs. those who reported not liking violent video games?; and, was there a statistical difference in helping for study participants who did not see the violent film in the context of those who reported liking violent video games vs. those who reported not liking violent video games?

    Answering those questions would seem to tease apart media effects from trait effects (state-by-trait effects as it is more commonly described in the literature).

  25. #25 Matt
    May 11, 2009

    I have to say that as person that likes games like GTA, runing people over with cars, In the game is fun but,when going home from school one day I saw a hit and run. What did I do? Pull my car over, call 911, and see if the lady was okay or not. I did not stop and think if I should help. I did somthing as fast as I could. She did not die by the way. I like movies and games like that. I can’t ever do that stuff in real life, so I can at lest play it in a game and/or see it in a movie or t.v. show. What about the news? I hear about rape, drunk drivers, war,and death everyday on the news. Can you say that “numbs” us?