Gender Bias

"WoW's like Texas.
Once you live there for a while, you always come back. Always" - Lara Crigger

A few weeks ago, I came back home to World of Warcraft. I'm not necessarily proud of this. WoW doesn't even make my top 3 MMOs of all time (Lord of the Rings Online, Neocron, Star Wars Galaxies). But no matter how often I stray, WoW has all the right ingredients to be addictive. Addictive in a good way. The combination of work, rewards, power and society are just right for scratching that particular itch.

WoW is just the latest in an endless rotation of serial addiction. For weeks at a time, occasionally months, I will get into a groove with a game. I'll think about it in the corners of my day. I'll bathe in the anticipatory light of when I will play next, and revel in the reality. This is a good thing.

Which is why I was particularly annoyed to discover a study last week suggesting that this wonderful fugue isn't the result of any predilection of mine towards gaming. It's just because I happen to be male.

Last week, some folks at Stanford University (who are undoubtedly smarter than I am) announced the results of their study on game addiction and gender. They took two groups of subjects, 11 male, and 11 female, and hooked them up to a functional MRI - a device designed to show brain activity in real time by measuring increases in blood flow in the brain . They watched them all play a game.

They didn't pick a real game like Tetris. Instead, here's what they came up with:

The researchers designed a game involving a vertical line (the "wall") in the middle of a computer screen. When the game begins, 10 balls appear to the right of the wall and travel left toward the wall. Each time a ball is clicked, it disappears from the screen. If the balls are kept a certain distance from the wall, the wall moves to the right and the player gains territory, or space, on the screen. If a ball hits the wall before it's clicked, the line moves to the left and the player loses territory on the screen.

Click on things as fast as possible before it gets too insane. I've played this before, I think it's called Missile Command.

They had these 22 folks play this game over and over again in 24 second bursts. They used a functional MRI scanner to track brain activity. And based on this, the findings are presented at Stanford Med's web site. This is not, I feel obligated to point out, what is in the Journal of Psychiatric Research. Somehow I must have let my subscription run out. Either that or my darn neighbor is stealing it off the driveway again.

In the interests of brevity, allow me to Fisk the Stanford article.

"The females 'got' the game, and they moved the wall in the direction you would expect," said Reiss, who is director of the Center for Interdisciplinary Brain Sciences Research. "They appeared motivated to succeed at the game. The males were just a lot more motivated to succeed."

This strikes me as a rather unscientific conclusion to draw. First off, I have a problem generally with small sample size studies. Yes, I understand that a carefully constructed small sample can have huge statistical significance. But even though I understand confidence intervals and Z-scores, I can't get around the fact that we are talking about 22 people here. How were they selected? How many of them play games on a regular basis? I can guarantee you that if I sit a non-gamer down with a "generic" reaction-time and mouse-accuracy game next to one of the Gamers With Jobs crazy-good Team Fortress 2 players, the non-gamer is just along for the ride.

Further, to somehow suggest that motivation in a game is somehow gender based solely because of this observation seems a bit of a stretch.

After analyzing the imaging data for the entire group, the researchers found that the participants showed activation in the brain's mesocorticolimbic center, the region typically associated with reward and addiction. Male brains, however, showed much greater activation, and the amount of activation was correlated with how much territory they gained. (This wasn't the case with women.)

Games are addictive!? Shocking! But the point here isn't that as much as the continued conclusion that males are more stimulated by success - meaning the better someone did at the game, the more excited this particular part of the brain became. On the face of it, this makes sense. They've already stated that in this group the men did better, and after all, if the little bell didn't go off in your head for each level-up, why would you keep playing? So why would it be surprising that these particular women would also show little involvement when you look at the brain scan?

The findings indicate, the researchers said, that successfully acquiring territory in a computer game format is more rewarding for men than for women. And Reiss, for one, isn't surprised. "I think it's fair to say that males tend to be more intrinsically territorial," he said. "It doesn't take a genius to figure out who historically are the conquerors and tyrants of our species"”they're the males."

And here, ladies and gentlemen, is our Fox News moment. We've gone from "we found 11 guys who are good at games" to the tired old saw of "men destroy everything" in one quote. No, I don't have a long list of female torturers and tin pot dictators to rebut with. But I do think that one-liners like this coming from the supposed cognoscenti do nothing to help the world understand the subtlety of human history, to say nothing of the world of videogames.

Reiss said this research also suggests that males have neural circuitry that makes them more liable than women to feel rewarded by a computer game with a territorial component and then more motivated to continue game-playing behavior. Based on this, he said, it makes sense that males are more prone to getting hooked on video games than females. "Most of the computer games that are really popular with males are territory- and aggression-type games," he said.

I have a feeling he started out trying to say "well, for this game ..." in an effort to not paint with a ginormous brush, but by the time he was done talking he'd put his foot back in his mouth.

What the heck is a "territory- and aggression-type game?" As far as I can see, this includes essentially any game that is not a form of puzzle solving solitaire, and even then I think I could survive a bar-stool argument. Freecell? All about claiming the stacks. The Sims? More stuff, bigger house. Every real time strategy game. Ever first-person shooter. Every arcade game in which you must press a button to keep the game alive can be considered "aggressive."

I can come up with a few exceptions. Phoenix Wright. Myst. Deduction and exploration games. But even Endless Ocean is ultimately about exposing more of the map - gaining more territory.

I'm not so naive as to suggest that there are not differences between men and women. I defy any parent of a matched-set to think that gender is not genetic. I'm also not going to suggest that games aren't often violent, aggressive, and hyper-stimulating in the same way as a high-energy action movie. But the way the results are presented here, and which will most assuredly be trotted out onto Fox News for years to come, are insulting. We know from sampling a much, much larger group of people that almost 40% of gamers are women. Through the sound bites summarizing Reiss' study, the general public will be led to believe that this 40% are not only "bad gamers," but somehow an anomaly.

Gamers, game developers and game journalists (myself included) are all complicit in this. Just as we fail to take our passion seriously enough to defend it, we also fail to debunk the very stereotypes this study is reinforcing.

Millions of women play games of all kinds. Let their experience of gaming - not that of 11 study participants playing lo-res Missile Command - speak for itself.

Comments

Quintin_Stone wrote:
SommerMatt wrote:

Uh... it *IS* Stanford.

As a holder of a BA in Psychology, I find such a casual "lay" (as in "person") dismissal of this study a bit baffling. Just because you don't agree with the findings doesn't mean it isn't a valid study.

Look, here's the problem: you don't know the history of psychology. I do.

Jeez. Stop being glib, Quintin.

KaterinLHC wrote:

Jeez. Stop being glib, Quintin.

As a someone with a BA in Creative Writing, I regret to inform you that the word "glib" was actually first used in 1599 to describe slippery white worms. I see the definition has not changed. (couldn't resist).

(Sometimes, like this, my OED is worth more to me than all the fMRI machines money can buy.)

rabbit wrote:

As a someone with a BA in Creative Writing, I regret to inform you that the word "glib" was actually first used in 1599 to describe slippery white worms. I see the definition has not changed. (couldn't resist).

(Sometimes, like this, my OED is worth more to me than all the fMRI machines money can buy.)

I see us engineers and you will English majors will continue remain the most bitter of enemies.

Here's my problem with this study as described in the article: they only played one game. How can you come to a conclusion on the type of game that appeals to a particular gender when the study only includes on type of game? That's... mind boggling to me. "The findings indicate, the researchers said, that successfully acquiring territory in a computer game format is more rewarding for men than for women." Then why not a comparison to a type of game where winning does not involve "acquiring territory"?

[quote=Quintin_Stone:|  I see us engineers and you will English majors will continue remain the most bitter of enemies.[/quote]

I also agree on your scale arguments Quintin.

I loved the article. Whether its 22 or 220,000 it says nothing about me. Thanks Rabbit.

Elysium wrote:
As a holder of a BA in Psychology, I find such a casual "lay" (as in "person") dismissal of this study a bit baffling.

As a holder of a BA in English, I find your criticism of this professional piece baffling.

Seriously, though, let's not deify scientists and assume just because they have long strings of adjectives near their name that they are infallible, unbiased or imperturbable. Julian isn't approaching this from an uninformed position, and his questions seem legitimate.

Good lord, people.

The point of mentioning that I had a background in psychology is that people were slamming this study as "bad science." I had to take several classes on experimental methods, and my only point was that if you don't know how psychological experiments are conducted, then YES, it may sound like a small sample, or "bad science" or whatever... but this is how studies are done.

No one is saying you can't find fault with the findings... I'm pretty sure my original post said that you really can't KNOW that unless you actually look at the data and methods. *I* haven't looked at them, so I'm not telling anyone to "shut up and take it." I *AM* saying to not jump on the bandwagon of criticism if you don't have facts to back it up.

I gotta say the tone of these responses has been about the snarkiest thing I've ever seen on GWJ... and from mods no less. Nice.

I gotta say the tone of these responses has been about the snarkiest thing I've ever seen on GWJ... and from mods no less. Nice.

Either the tongue-in-cheek nature of my previous comment fell short, or ... well, let's leave it at that.

Still, pot meet kettle. And, I mean that one sincerely.

Elysium wrote:
I gotta say the tone of these responses has been about the snarkiest thing I've ever seen on GWJ... and from mods no less. Nice.

Either the tongue-in-cheek nature of my previous comment fell short, or ... well, let's leave it at that.

Still, pot meet kettle. And, I mean that one sincerely.

I don't see any "snark" in my OP... I was trying to raise a point not to rashly dismiss things without looking more closely at the data and methodology.

As you said, I guess I'll just leave it at that.

As a fan of insta-memes, I approve of this thread. You have given Gears n Beers weeks of material. Bless you.

Elysium wrote:
I gotta say the tone of these responses has been about the snarkiest thing I've ever seen on GWJ... and from mods no less. Nice.

Either the tongue-in-cheek nature of my previous comment fell short, or ... well, let's leave it at that.

Still, pot meet kettle. And, I mean that one sincerely.

I believe I sir have a trademark on that type comment.

With all the developed territory Jimmy Carter is responsible for, his mesocorticolimbic center must be going nuts.
Also, now it's my flippin' point!

I have to agree with SommerMatt on this one. While the conclusions seem to be a presumptuous, without examining the actual study we have no way of knowing if the actual "science" of the study is valid and being that it was conducted at a major university, we can only assume that it is. Not to mention the fact that at the end of the formal written study there is probably a portion explaining what their shortcomings may have been, how they could improve on them and what to look at in further studies. Yes, the sample size is small and I guarantee that is something mentioned in the discussion portion as well as a suggestion to re-try the study on a grander scale. It is a small study that will probably jump start larger and slightly altered studies to look at the same concept. It should not however, be completely discounted simply because the sample size is small.

That being said, I do feel that the study had several flaws and the conclusions are weak but as I said, it does give the Psychological world fodder for further investigation.

With all the developed territory Jimmy Carter is responsible for, his mesocorticolimbic center must be going nuts.

He is a peanut farmer!

Psych wrote:

I While the conclusions seem to be a presumptuous, without examining the actual study we have no way of knowing if the actual "science" of the study is valid and being that it was conducted at a major university, we can only assume that it is.

No, no no no no no no.
Never trust because of credentials alone. A scientific work must be judged for merit based upon how it collects and analyzes data, not by name alone.

I was merely responding to the attempt to pad out your comment with your cirriculum vitae. Which is why I replied with my relevent credentials to point out that many of us here are plenty able to discuss the scientific features of this study and it's ilk. The level of snark in my comment was in as direct a proportion as I can make with the level of snark I percieved in yours. I'll try to keep this more civil.

I do know how studies are done. And I did look her up on the APA website and Google before I made my comment. The first author's name is Fumiko Hoeft. This study isn't listed yet, but while you're there you can also look at her previous study structured the same way where she correlates fMRI results to a kid's future reading skills (PDF) by comparing dyslexic and non-dyslexic kids. Here's the APA PSYCInfo precis.

But even if you don't know how to do that, anyone can see some common problems. Just using the prima facae facts:

-- 11 subjects is not statistically signifigant when you're trying to extrapolate to half the human race. Heck, I wouldn't extrapolate that as far as the line down at my local Starbucks.

-- There is no comparison to other formats of stimuli that would also trigger that sort of response. Would it be different if it was a physical board game? How about a trick-based card game where the goal is to get your opponent's cards?

-- There is no correction for other gender-based influences. There are studies that show that there are differences between the way the genders interact with electronics and computers in general. Are these results merely a reflection of that? For example, there are user interface studies that indicate the genders respond differently to color, shape, and movement on a computer screen. At the least you'd want to try your test with a couple different screen layouts to make sure you haven't run into something like that.

-- There are no other correlation behaviors tested. There are many other ways the concept of territorial ownership is played out in society; none of them are looked at.

And the big elephant in the room - there is not nor has there ever been any causal linking of any behavior of any sort to general or specific fMRI results. Certainly not to something as complex and nebulous as addictive behaviors with a non-physical source. fMRI is imprecise compared to other imaging techniques, and no one has any idea how blood flow to the neurons relates to the exterior behavior of the subject in the real world.

If this was an entry prospectus or a technique test as a prelude to a study with a scale and breadth comensurate with what they claim they're trying to find out, that would be one thing. But it's being published and treated as if it was finished work with concrete, applicable findings. This isn't some student, either. She has credentials and and because it's being published, people are taking that tiny little flicker of orange she's reporting and carrying it way way too far. Just like it's many compatriots that are being used to bludgeon us in many aspects of life.

It didn't use to be this way, because this stuff required hardcore effort to find. Used to be you had to pay $2000 a year or more to have access to these journals, or it was a hard slog to go down to the library and a lot of work to find them. Nowadays anyone with Google and a short deadline can dig in there and find something to support their query letter or editor's instructions with a few taps on the keyboard. And then people who aren't paying attention and just trusting in the authority of the press or the credentials of the quoted materials take it as truth.

"That's how studies are done" is a problem, not a feature. It IS bad science. Just because she's not the first one or the only one to do it doesn't make it good science. As the study is constructed it's scientifically questionable, and when it comes to discussing behavior it's misinformation at best. The irresponsible publishing of studies like this is a problem that is having ripple effects in our real lives. And anyone willing to do a modicum of looking into it can see it.

All I was saying is that based on limited info and without actually going out to PsychInfo and finding the article, we have no way of knowing whether the study is scientifically sound. I understand that credentials are not everything and as I said, I do take issue with the limited scale (among many other things) however, since I was too lazy to go read the study myself, I could only go off of what the casual conversation had presented and one of those things was the University at which the study was conducted.

Rabbit wrote:

In the interests of brevity, allow me to Fisk the Stanford article.

So you learn something new every day. I am not a blog guy, I read the occasional one here and there, but in no way am I part of the blogosphere. So I read Rabbit's article as I usually do (completely but not going to the external links). I enjoy the writing and rarely feel a need for more or to go to his sources. So with all the back and forth in the comments I reread the article and actually clicked on the links.

Funny thing, if you click Rabbit's Fisk link, read the definition on it in Wikipedia and then look at his article, your overall view might change. In fact, I think some of the back and forth might not have happened based on a better understanding of where Rabbit was writing from with regard to his Fisking of the Stanford article.

Just my two cents. I have enjoyed reading the back and forth but felt that others might have skimmed over the Fisk term as I originally had.

I feel the same way about that old text MUD Gemstone IV, it keeps sucking me back to it.

I just fell a bit in love with momgamer after that.

And not just a elementary school, 'fell for the teacher' crush, the bad, heart-felt, "i'll even do the dishes, you can keep the remote control" kinda love!

Momgamer, that was a nice piece of writing right there.

Elysium wrote:
As a holder of a BA in Psychology, I find such a casual "lay" (as in "person") dismissal of this study a bit baffling.

As a holder of a BA in English, I find your criticism of this professional piece baffling.

As a frequent holder of females, I think you're all silly. I especially want to point out that, having read the article Rabbit links, the study was rather clearly focused more on gender differences in territorialism than in videogame addiction, with the latter being more the emphasis of the pseudo-journalist who wrote the article.

That having been noted...

TFA wrote:

The men, however, wound up gaining a significantly greater amount of space than the women. That's because the men identified which balls"”the ones closest to the "wall""”would help them acquire the most space if clicked.

...

Male brains, however, showed much greater activation, and the amount of activation was correlated with how much territory they gained. (This wasn't the case with women.)

It seems to me that one could argue from this that men are more likely to take larger risks in favor of larger rewards.

rabbit wrote:

As a someone with a BA in Creative Writing, I regret to inform you that the word "glib" was actually first used in 1599 to describe slippery white worms. I see the definition has not changed. (couldn't resist).

(Sometimes, like this, my OED is worth more to me than all the fMRI machines money can buy.)

The full OED costs roughly the same as an fMRI machine, doesn't it?

$251 shipped.

Hi everyone, I'm a long time reader, first time poster.

After reading the article and responses I felt somewhat motivated to add my own 2 or 3 cents, so thanks for the inspiration to get off the bleachers.

As something somewhat relevant to the conversation (a ploy I'll use to set up the expectation that my reply might be informed and/or worth spending some energy reading), I've got me one of those PhDs in Cognitive Psychology, and a doctoral minor in Quantitative Psychology (statistics) from NYU. I've done some training on the physics / operation / statistical analysis of MRI / fMRI at the Nathan Klein Institute up in Westchester, NY. And I've been playing games on computers and consoles since the Atari 2600 / Atari 400.

There is a pre-press version of the *actual article* online here: http://spnl.stanford.edu/publication...

It's great that many are skeptical of the conclusions and methodology of the article, if only for the reason that informed skepticism is healthy and necessary for Science!

I've read some comments in the article / replies that give me pause, though.

I've always seen a lot of criticism of studies like this, people knocking others credentials, methods of inquiry, findings, etc. And reading comments from authors (to the press, outside of peer reviewed articles) I completely understand why. Authors make provocative statements about a study to the press that they would be hesitant to put in a peer-reviewed study. And it's those statements that people react to.

But sometimes the response to those comments in the press goes overboard. Just because the author that made the comments used MRI / fMRI in the study isn't a reason to indict the use of MRI / fMRI as a method of inquiry. I take some issue with some of momgamer's comments [sorry momgamer ]...

Example... there is not nor has there ever been any causal linking of any behavior of any sort to general or specific fMRI results.
If you are saying that activation in various regions of the brain has not been shown to have a relationship with behavior, that is not true. There are different methodologies for doing this (one of which I took part in at Stamford doing some research at Yale) - for example lesion studies. Cognitive deficits can be localized in the brain due to special patient populations such as those who have some sort of physical insult to the brain (lesions). If a patient has had a very localized trauma (not a diffuse injury like blunt force trauma) and experiences an demonstrable cognitive deficit due to the injury, then an MRI can be useful in showing that the affected region is at least partially responsible for part of the cognitive process(es) that are impaired. Temporary effects similar to lesions in localized areas of the brain can be induced in animals as well for similar studies. Furthermore, certain drugs can cause localized effects in the brain (for example, there are drugs that have a selective effect on the amygdala) to do research on neural correlates of emotion and memory, as an example.

Example, fMRI is imprecise compared to other imaging techniques. This is somewhat true, but as a general statement is misleading. EEG studies have better spacial resolution in most cases than most fMRI studies. But theres a lot of factors at play here. Without going into the physics of it, there is a trade-off between spacial and temporal resolution when doing an fMRI study. Resolutions are dependant on how big of a magnet you've got (the study mentioned here used a 3T MRI which would have been decent back in the day but is now a bit lightweight imho). But we are talking millimeters here in terms of precision, not completely different cortical or subcortical structures. If they say they see a difference in activation in the amygdala, I wouldn't doubt it. As a general argument, I wouldn't call fMRIs imprecise and if one wanted additional precision in localization, one can do both an EEG and fMRI.

I'd suggest it is bad to indict all of science because of a description of one or a handful of peer reviewed articles, saying... "That's how studies are done" is a problem, not a feature. It IS bad science.
Future *responses* to this article are "science" too - its a series of articles, knowledge built on experimentation *over time* that is "science". It's a *process*, not a single event. The process allows for the formation of hypotheses that *turn out to be wrong* so that there are no constraints in the method of inquiry. Otherwise, it ends up being politics, not science.

A few general comments...

Small sample size in and of itself isn't a criticism of a study. Small sample size just means there has to be larger differences (relative to the variance) for something to be statistically significant. There are lots of psychophysically flavored studies that do fine with very low samples (like 5 people) because they make use of LOTS of trials. These kinds of studies are accepted during peer review (look to literature in perception) because it's assumed they are measuring common physiological / psychological functions or phenomenon. As an example, it's assumed that certain processes (for attention / perception) operate in the same way for all people (that are not suffering from some cognitive deficit). All that to say that you can get a statistically significant result from a low number of participants.

But... statistical significance is not the same as statistical validity. The arguments here are more about the validity of the study (is it testing what they claim? etc.) and whether the results are generalizable. Have the authors found a gender difference in reward response to video games? Uncorrelated with prior gaming experience? Applicable to all video games? Etc. Skepticism about the validity (specifically, generalizability) of the study and its conclusions is warranted and needed (in this as with any other study).

Specifics from the study...

Twenty-two healthy Stanford students (19–23 years) participated in the study (11 females). ... Results from a survey showed that there were no gender differences in the degree to which subjects played computer/video games or utilized a computer in everyday activities (P > 0.05).

We excluded possible confounds due to neuropsychological profiles, lower-level motor performance, and computer and video-game experience.

While the lack of significant difference in computer/ video game-play between genders suggest that motivational level were similar between females and males, studies that tease apart gender differences from possible confounds such as difference in motivational states are warranted.

To our knowledge, this is the first fMRI study of computer video game playing examining neural systems related to reward and addiction. We show novel evidence that gender
differences exist in the mesocorticolimbic reward system during an implicit space-infringement task with greater activation and functional connectivity observed in males than in females. The overlap with neural processes, underlying addiction may help us understand the greater propensity of males for playing video games in a repetitive manner.

---

My take, for what it is worth. I'm glad people are investigating video games and psychological processes. It's interesting to me in general, and I enjoy reading anything about video games. If the finding isn't valid, I'm confident the error will be discovered over time. I think it's an interesting study, but I think the (out of article) comments by the authors seem irresponsible (being highly speculative). There is a limit on generalizability on the study due to the fact that 1) the "game" had implicit rules, 2) it was one type of task, etc.. There are potentially confounding factors that are were not included in the investigation (understanding that this was a "game", motivational factors, gender of experimenter and desire to please or impress, etc....). I find myself just as skeptical (especially since I got started in psychology studying gender differences in math anxiety) as most seem to here, but I pause before indicting "science" and its conceptual (statistics) or physical (MRIs) tools.

rabbit wrote:

$251 shipped.

I suppose this is where you bring out the "size doesn't matter" shtick.

It covers more territory, and we all know how males like that!

NYTimes picked it up.

Here I thought you meant they (rightfully) linked to Julian's article. Jerk.

Certis wrote:

Here I thought you meant they (rightfully) linked to Julian's article. Jerk.

You should change his tag!

All About Mahni:

I'm afraid of the new guy. He's a doctor Jim. It's like he knows what he's talking about.

wordsmythe wrote:
rabbit wrote:

$251 shipped.

I suppose this is where you bring out the "size doesn't matter" shtick.

It covers more territory, and we all know how males like that!

Who cares, the OED keeps getting bigger because we keep pumping it full of words. I doubt it should be used as anything more than knowing what something means not whether a word is correct or not. I'm coming up with a new term right now, zenting , venting by just doing and it'll leave you with minty fresh breath!

Mahni,

I want to thank you, deeply, for de-cloaking and weighing in. I take every one of your comments and arguments seriously, and you've made me walk through the article one more time. Your distinction between significance and validity is an important one, as is the issue of underlying science vs. actual deployed methodology.

I'll hope you keep uncloaked. Great ideas here.

J

PS - Zen, hard to see the location of tongue in your mouth, but I can assure you as someone who spends 10 hours a day trying to get ideas across to vastly different kinds of audiences, being able to walk over and read about a word can be a powerful and stimulating process. It's not understanding that the color "red" describes a particular spectrum. It's reading about how the word has been used over time, perhaps extracting a few quotes where the usage was changed, suggesting alternate ways of describing whatever it was I was trying to describe. I see it as similar to looking at code samples to understand how, say, a conditional syntax can be constructed in an elegant way, instead of just reading the man page for "while."

I don't spend a lot of time looking up words like "red," but I do spend a lot of times using it as a thought-prod.

I have to agree rabbit, Mahni's differentiation of significance and validity is is a very key point. A lot of the back and forth actually hinged on this relationship and the causal effects or is it affects (reminder buy OED) between the two concepts. Thank you Mahni for weighing in, you have shed quite a bit of additional light on this topic..