A new piece of research has been published detailing findings that show a correlation between playing video games and a lesser ability to focus.
The study conducted at Iowa State University took 51 men aged between 18 and 33, recorded their gaming habits and then observed their brain as basic tests were conducted. The results showed that “high” gamers (with an average of 43 hours a week spent gaming) had noticeably different brain patterns than “low” gamers (with an average of just under two hours of gaming per week).
Whilst I don’t pretend to understand the majority of the scientific results in the paper, they do back up the following statement taken from the Discussion section of the article.
Together, these data may indicate that the video game experience is associated with a decrease in the efficiency of proactive cognitive control that supports one’s ability to maintain goal-directed action when the environment is not intrinsically engaging.
Before angrily leaving a comment about how correlation does not equal causation, the report acknowledges this openly and calls for more study to be done, building upon these findings. The press release sent out with the study however opens with: “Parents have long lectured their children about the mind-numbing effects of playing video games all day…” Great.
Talking to Negative Gamer, Robert West, one of the co-authors of the study, acknowledged that it’s unknown “whether this influence is causal (i.e., does playing the games cause the effect on attention or are folks who play some types of video games less likely to use proactive attention)”. Adding that “there could be different effects of various genre’s of games (FPS vs RTS). Again, more research is needed here.”
“leaving a comment about how correlation does not equal causation”
Damn…
I still don’t like how they word it though, and only 51 people? Would that even qualify as an good undergrad project? :\
“Before angrily leaving a comment about how correlation does not equal causation…” – Fuck that, it’s an obvious flaw in the assessment; just because it undermines your both that and your article doesn’t make it any less important. Furthermore, a “study” on 51 people is a joke, call me when it’s performed on 10,000 or more.
@rock: Now, not that I’m saying you didn’t read the whole article, but if you read the whole article you would see that the report simply says there is correlation and calls for more research to be done to find if there is any causality. Neither my, nor the researcher’s “point” has been undermined. If anything it’s been re-enforced by your comment :P
TLDR
@rock: Actually, a study on 10,000 or more people would be a bigger joke. Statistically speaking, there is no benefit from increasing the sample size beyond a thousand or so, so it’s just a waste of time and resources. You also run into the problem where, when you start increasing N dramatically, trivial differences become statistically significant. This makes it MORE likely that they find correlations. There is nothing wrong with having 51 people as a sample size, assuming they randomly chose the 51 people. Just FYI.
@rock: I should clarify; there’s nothing INHERENTLY wrong with having 51 people in the sample. The fact that they found a statistically significant correlation means that the difference between heavy gamers and light gamers was large enough to be distinguished despite the low power of the test. If they hadn’t found a correlation, then it might make more sense to increase the sample size, because then the difference between gamers doesn’t have to be as large. And again, if the sample size if large enough, the difference could be minuscule and they’d still find a correlation.
To those complaining about having only 51 participants in the experimental study.
Newsflash: That’s considered a lot in most experimental paradigms. No experiment ever had 10,000 people… I mean EVER.
Experiments aim for internal validity, not external validity, so they tend to use fewer subjects than, say, a survey.
Again. 51 people is a lot.
Obviously we’d have to check the paper to see what kind of reliability and validity measures were used, but you can achieve a high degree of validity with 51 people.
Learn about how scientific experiments are actually conducted before trying to pick studies appart. It keeps you from looking uneducated.
Here’s your homework. Go look up how many people participate in experiments (not correlative studies) in Science and Nauture.
I love the smell of denial on a saturday morning.
The argument makes sense. Sometimes logic is enough, I would say it is generally the personality types but then it’s just a statement. You’ll continue to be a gaming addict (note the word…addict) regardless of what people say, other addicts support your addiction
Or maybe people who have shorter attention spans are drawn to video games while people with longer attention spans aren’t.
Also the study states that there is a reduction in proactive cognitive control. I’d appreciate if somebody could explain what exactly that means because google yields nothing other than an explanation of cognitive control. And looking at cognitive control (basically problem solving), I don’t see how that could relate to the Stroop test which is basically testing reaction time.
Finally, they did this test specifically on gamers to see to test their attention spans. This is one of those witchhunt studies that garner attention and funds but really just fuels a misunderstood fire. Why would you do a study on gamers’ attention spans that could only produce correlative evidence? There’s already a bunch of studies that say the same thing, but don’t prove anything other than “there needs to be more research.” Why not piggy back off other research and do the research that needs to be done.
@derp
Well, tbh I don’t particularly care that much, but after a little bit of digging I did find that only using 50 participants in a study gives a margin of error between 15-20%. Now, I don’t know about you but that’s a pretty large margin of error. If you’re comfortable with that, then it would seem to me that you’re a very, very, very shitty scientist.
Although I’m appearing as uneducated, so what do I know? <3
TL:DR.
when we (technical university) make user studies on user interface prototypes or augmented reality concepts, we always have 30 participants. This is the “minimum accepted” amount of participants for the science community. Also, the test covers always three or more different aspects – but in reality, only one aspect is valued. The other aspects are setup in a way to further enforce/clean up the “valued” results.
To further assure the accuracy of the study, the questionaire and/or test setup are convoluted and redundant to a point where the testsubject can not really “know” what is tested for exactly (to prevent him gambling the system).
15-20% margin of error is actually a very good value if connected to a binary research result. If the answer you try to find is “yes or no” (which this paper does), every value with more than 70% certainty is good.
@Brandon: i like your point (witch hunt) very much. Because there are some studies out there that test capabilities not only within the gamer population, but gamers vs. notgamers. And most of these studies paint a very interesting picture IN FAVOR of gaming. But those get cast aside by the catchy “violent games makes people slaughter kittens” variety, because nobody likes to admit that they were wrong. The label of games being bad will not go away because of that even though there are studies out there that say otherwise.
30 is an “minimum accepted” and 15-20% margin of error is very good?
… *cries*
@ScottyGrayskull: for a test that tries to make a BINARY STATEMENT, that is.
So if i want to get a yes or no, even a 49% error margin would be sufficient, strictly viewed.
And that sample sizes are that “small” has practical implications:
for instance the last study we made tried to find out why people have a tendency to move the mouse to the right for about 200 ms in a special screensetup. The test itself took about 20 minutes per person. And then there was the questionaire: 40 questions were to be answered. Traditionally, those are filled out on paper (to warrant absolute anonymity). The researcher then has to analyze the questionaire. Thats about 15 minutes per questionaire. Now bump up the sample size to lets say 100 people and the researcher will NEVER meet his deadlines. (Research is way harder on deadlines than game development – if you are 5 minutes late, you will not get published and loose about 3 months of work)
It wouldn’t really surprise me if this was true…at least in some small way.
Can’t say I care a whole lot though.
I’m too busy being distracted.
@NoZart: Not to harsh on your parade but quantifying brain function is hardly a binary statement. There must be acceptable boundaries for how they define a significant loss in focus and brain function.
But this really doesn’t prove anything because of the small sample size. He recommends more studies so that they can expand their pool of data and see if the small test supports a larger subject group. The margin of error along with the small test group could suggest that there is a contingent of subject that just have short attention spans.