« Castronova on NPR | Main | Syndicate Sponsorship »

Jan 27, 2004

Comments

1.

So I was going to say "yes, it has been looked at quantitatively, see Nick Yee's page." But clearly, you mean more than what you have done. :)

There's a pile of studies on genderbending in particular, but also on species/race choice, from back in the mud days. I don't have any pointers handy off the top of my head, but Lydia's collection at

http://www.godlike.com/muds/

has links, so does the archive at ibiblio.org

http://www.ibiblio.org/dbarberi/papers/

Elizabeth Reid Steere's work probably has some relevance.

2.

I hope TL will get a chance to weigh in on what she might like to see quantitatively (or maybe references).

Personally, what I would like to see explored further is Amy Bruckman and Joshua Berman's Turing Game experiment. Gender and age of one person was explored using a twenty questions format. The idea was to assertain the true gender and or age of a person even if that person was assuming a gender or age that was not their own. They used only 10 subjects, but I would like to see something similar run over a longer period of time with more people.

The Turing Game: Exploring Identity in an Online Environment PDF
Berman, Joshua and Amy Bruckman. "The Turing Game: Exploring Identity in an Online Environment." Convergence, 7(3), 83-102, 2001.
http://www.cc.gatech.edu/~asb/papers/convergence-tg-01.pdf

3.

Back in early UO days, I had a 2nd character that was a female. Initially this character is what some call a 'mule': I basically used the avatar to mix potions and that was about it. I started messing around a bit with this character after my primary character was maxed out, and eventually this character turned into my primary character for a while.

One of the main reasons was that a lot of the 'k00l dudz' and other players that I'd consider somewhat immature treated this avatar much better than my male avatar. I was into the fighting/guild wars/pk stuff, which seemed to be where a lot of the younger players spent their time. This led to a lot of trash talking, rez killing, etc. This rarely hapenned when I would use my female avatar, which was always interesting. If someone would ask me what my 'real' gender was, i had no problems telling them I was a male. Surprisingly enough not many people even asked this question.

Just an observation from the field...

4.

Nick Yee: So what happens if you switch the genders of certain users in an MMORPG so that others perceive them as the opposite gender but the players see themselves unchanged?

What exactly would it mean to "switch the genders of certain users"? Switch the appearance of their avatars? Switch a sex flag in a publicly available character description information? Particularly among experienced players, I would guess that communication content and actions play a large role in sex categorization of players in MMORGs. Empirically separating different senses of character in relation to sex categorization is probably impossible, because of the intimate intertwining of sensory modes in making sense. See Section I of "Sense in Communication," at www.galbithink.org.

Scott Moore: Personally, what I would like to see explored further is Amy Bruckman and Joshua Berman's Turing Game experiment.

I agree. The paper points in an interesting direction. But the abstract is rather misleading. "Do men and women behave differently online?" is a dull question. The paper actually addresses more intellectually challenging issues: the scope of common knowledge and persons' ability to execute complex strategies. As research on general characteristics of human beings, the experimental set-up suffers from obvious selection bias. A better-controlled experiment with an appropriately chosen sample size would be great work.

Nick Yee: Would we be able to get male users to act more sensitive and considerate and more people-oriented?

Given male-female differences in school performance, college graduation rates, suicide rates, and incarceration rates, I can see your rationale for seeking to help males. But it seems to me that you might consider more explicitly and carefully males' real-world life opportunities. The latter, in addition to being what should be an important public concern, might also relate to online behavior. What, for example, are the implications of continuing male-only US selective service registration for the prevalence of male "killer" types in online worlds?


5.

As anecdotal evidence, I agree with Bart that when I play female characters there is a slight but noticable effect on how other players treat me vs. playing males. People are a bit 'nicer' to the girl and tend to give more free things to her. I've found advice and help given to be roughly equivalent. And, speaking broadly, most players are more casual and/or rough-and-tumble with the males. Basically, people seem to treat avatars under the same gender conventions as they would a RL human. This is a generalization, however, and it may hinge on the reality of a majority of players using avatars of the same gender (I don't have any numbers to support that, but I think it's a fair guess). An interesting thing for me is that once I'm playing a character, I don't (usually) actively consider my avatar's gender until another player invokes gendered bahavior toward me; eg. batting her eyes at my male, or putting the mack on my female.

"Surprisingly enough not many people even asked this question."

I've found the same thing, but I'm not surprised. People want the suspension of disbelief; if they were really concerned with the player's qualities over the avatar's, they wouldn't spend a lot of time in a MM "game" world. Social worlds are rather different.

As for the social engineering angle, I don't think you can bootstrap the entire system toward utopia with the Pygmalian effect. The cost of "get[ting] male users to act more sensitive and considerate and more people-oriented" would be to have females act more typically male. No net change to the system, just some identity confusion :) As an academic point I'm sure it's possible to demonstrate an effect, but I doubt it would be otherwise productive. [Specifically, this proposal is a more subtle version of, "Will Player X respond angrily to Player Y if everything that Player Y types to Player X is secretly changed to a string of curses and insults?" Of course--but this doesn't have any novel application to RL, or even enjoyable gameplay]...Although it's possible that VWs in all their as-yet unseen manifestations could push society toward a healthy psychological androgyny, over time, by allowing heretofore unavailable perspectives. Someone start a 50 year longitudinal study...

6.

The Looking Glass Self issue isn't how others treat you differently when you appear different (gender, race, age etc), but how their treatment of you changes you and your behaviors.

Douglas: I would guess that communication content and actions play a large role in sex categorization of players in MMORGs.

Yes, but when A interacts with B as if B were female, how does this change B's behavior? Does the communication content and actions of B change because of A's interactions towards B?

Euphrosyne: Specifically, this proposal is a more subtle version of, "Will Player X respond angrily to Player Y if everything that Player Y types to Player X is secretly changed to a string of curses and insults?" Of course--but this doesn't have any novel application to RL, or even enjoyable gameplay

Think about it on a different level. Could we lower the overall Grief/Killer tendencies of a particular server by doing this, thereby creating a larger Socializer cohort? Could we draw in a larger female crowd into a game by making the overall atmosphere less achievement/competition-driven? Could we help underprivileged minorities overcome learned helplessness attitudes by placing them in simulations where they "wear" a different skin tone?

As more and more of our interpersonal communication revolves around technologies that allow manipulation of the Looking Glass Self without letting the user know, could we use this to affect the stability, development, or even balance of virtual communities?

7.

Euphrosyne: Of course--but this doesn't have any novel application to RL.

Classic Pygmalion Effect - When teachers are told that randomly chosen students have higher IQs and learning potential, those students learn more and score higher on IQ tests than other students in the class. Understanding these effects have tremendous RL consequences and applications.

8.

First off – this in not a phenomena of “EQ-clones”… this is an effect that has been going on as long as there have been chat rooms.

“So what happens if you switch the genders of certain users in an MMORPG so that others perceive them as the opposite gender but the players see themselves unchanged?”

If the visual perception of someone is female (name or avatar) then both males and females will be inclined to treat that player as a female until it’s proven otherwise. Even those who are jaded by strings of “female impersonators” will still subconsciously lean towards treating the individual as a female from the unconscious hope that the person really is what they seem. Naturally there are exceptions but by and large this is how it has always been and always will be.

“People act differently towards men and women, so how does this affect self-image and how one should act towards others?”

This will depend greatly on the player and their reasons for choosing a female avatar. As you said, not everyone picks an avatar as a representation of themselves. Some pick an avatar based upon what they like to associate with (or look at). And some pick avatars for other reasons. If the avatar picked is a representative one then the male who picks a female avatar already has some issues with their sexuality. Having other players (who are naturally predominantly immature males) treat them with special favor will likely be an affirmation for the troubled individual and make them feel better about themselves. Those who pick an associative avatar will, like those who choose the female for other reasons, will most likely simply find the attention amusing.

“Would we be able to get male users to act more sensitive and considerate and more people-oriented?”

No. To do this you’d have to have every player in the game appear female to them at which point the male user you’re trying to manipulate will just assume everyone is male unless they have a female name. If you give everyone a female name as well then they will just assume everyone is male. Beyond that, why would you want to? You want to write a game that improves society by raising people’s children for them? Even if you did it wouldn’t work. The males treat the female special because they perceive them as female. Once that perception stops then they revert to treating them as they would anyone else. So once they shut off the game an re-enter RL they will continue in the same behavioral pattern... those perceived as female being treated with favor over those who are male (though more likely with fear instead of favor due to the lack of anonymity in RL).

Try making a game that’s fun (an appropriate death penalty, minimum downtime, no grinding)… then worry about making a game that saves the world.

9.

Nick,
I understand the looking glass/Pygmalion effects and think that your goals are noble, if perhaps a bit misplaced. My example specifically responds to your proposal to basically alter a player's avatar randomly and without their knowledge or permission.

"As more and more of our interpersonal communication revolves around technologies that allow manipulation of the Looking Glass Self without letting the user know, could we use this to affect the stability, development, or even balance of virtual communities?"

I'm all for implementations that let players personalize their worldview. Electric rose-colored glasses are gonna be great. But your above statement is social eugenics, and unworkable at that. How many players will choose to play a game where they never know how others see them, and where what they see is false? There's a world of difference between allowing a player to customize views, and doing it centrally for them without feedback loops. Personally I find it more fruitful to consider why people chose their avatars as they do (your first question) than to entertain notions of negating those choices for them.

(For what it's worth, I just had an image run through my head of Ragnarok Online populated entirely by female avatars. An amusing option that I might toggle on for a bit if I could, but would quickly tire of. I think I'm going to go have an existential crisis before it becomes completely passe...)

10.

Nick Yee>So what happens if you switch the genders of certain users in an MMORPG so that others perceive them as the opposite gender but the players see themselves unchanged?

When players play characters different to themselves, this can influence their real self, yes. However, it can also affect the character. Much of playing a virtual world involves making tiny adjustments to your self and to your character until eventually the two align. It's as if there's a dialogue between the two, the resolving of which affirms the player's sense of identity.

The main worry I have about making people look different depending on who's doing the looking is that it means players are operating to a false premiss. I don't particularly care whether overall attitudes to men and women are gradually affected by the build-up of impressions gained from false presentations, but I do care if an individual is screwed up by them.

If a player thinks they're being treated nicely/nastily for one reason whereas actually it's for another reason, this is fine so long as they don't find out. When they do find out, though, it could have severe effects on them: the edifice of the personality they were building for themself was constructed on bad foundations.

The moment they're hit on by someone who they think is of the same sex but who is actually of the opposite sex, they'll figure something is wrong. In a virtual world that has gender, how are you ever going to stop them from finding out that their character's gender in other players' eyes isn't what they think it is?

Richard

11.

Here's my two cents on this topic (one of my favorites, btw).

Nick Yee> "the question of how gender, age and personality intersect with avatar creation and use"

Throw in motive and I think you would have an excellent study into human behavior. Even more importantly, you would have that study in the context of virtual worlds. If VWs become more and more prevalent and widespread, the study should be able to accurately predict population trends and characteristics, not unlike Nick Yee's Norrathian Scrolls.

Even so, I'm not that interested in the demographics. Personally, I think that were someone to conduct a study, one tact might be to take a sample of players from across the board (different types of gamers, differences of age, sex, race, etc.) and run each participant through a gauntlet of pre-made chars, i.e. a white male, a white female, a black male, a black female, etc. Then, rather than looking for demographics, examine what each individual has learned from the experience. How has it affected them, if at all, or changed their mode of thinking?

Personally, I've played as both male and female chars. The primary reason I chose my first female char was for roleplaying reasons. The reason I chose my second female char, with the same name, was for character identification. (I began identifying and became known by my first female char so I used the same name and sex again. It became familiar and even comfortable.) There are other reasons too but that's the primary one. Escapism, (not char stat advantage but) char customization, and other factors exist but they're all secondary in my mind.

Have I learned anything from my experiences? Yes, a great deal. Probably the most overriding item is the one I knew before I ever went online, men and women are treated differently. Online, it's by perception. Nothing too shocking there, despite the feminist movement etc.

It's somewhat interesting that game designers inherently treat the sexes differently themselves. Perhaps that could be an interesting area of study, what avatar characteristics and attributes say about the people who design the games and their thinking. (Especially as these are the people who are directing what the next games look like.)

I'm actually surprised by people who do not take every online perception at face value, with a few grains of salt. The same people who know better than to believe every advertisement are willing to automatically assume that avatar representations are at least somewhat reflecting real life. When I'm online, I try not to assume anything. I do make assumptions, often based on behavior rather than appearances, but I always try to remember where I am and that what I see is rarely what I get (so to speak). The break down in this logic occurs in highly cutomizable avatars, such as those in Second Life, where a portion of the population attempt to recreate their real world self in-game. I wonder how many actually do so to some degree of accuracy but I can see it in some of their efforts and real life photos in their profiles. (Assumptions galore here, I know.) I wonder why people try to do this? (Or even if they do?) Maybe this added complexity (the added choices and possibilities) as compared with the avatar choices in a game like EQ, DAOC or SWG provides additional insite or information?

12.

Alan> "When I'm online, I try not to assume anything."

This is intriguing because it's largely the opposite of my gameplay approach. I tend to try to assume everything, so to speak, and take the game logic on its own merits. If a female character treats me in a stereotypically boorish, 'male' way, I'm far more likely to think "that bitch..." rather than "what a jerk...".

My impression is that younger players are growing up in this environment and building more flexible internal models of self and reality. They can switch between disparate realities without the compulsive need to cross-reference them back to a single root "reality". More of an ungrounded associative network than a heirarchical tree.

That said, there will always be people who try to faithfully reproduce their RW selves in every virtual environment. But with social communication becoming increasingly mediated and malleable, I wonder how long the persona of the flesh will be considered more authentic than a well-crafted, persistent avatar? I, for one, hardly care what the gender of a player is if I'll only ever interact with their avatar, and their avatar has a consistent personality. With players free to selectively express and project portions of their RW selves into partitioned realities, it might be a losing game to try and classify it all in the RW under the simplistic Freudian dichotomy of male/female.

13.

"When teachers are told that randomly chosen students have higher IQs and learning potential, those students learn more and score higher on IQ tests than other students in the class. Understanding these effects have tremendous RL consequences and applications."

That's fine but it's not exactly applicable. The above example the teachers are instructing the students they think are different in a different way. Teach a kid as if he's smart and he will become smarter (or fall behind). It didn't alter the way the teacher taught other students even those in the same class.

To carry that over to your hypothesis, even if you could convince a player that certain other players are female then only those other players are going to be treated differently. The players whose view is being altered will not carry that interaction over to other players he perceives as male, nor to anyone else outside the game.

The flip side of this example is that by being treated as females these other players would, theoretically, begin acting like females but I think that depends on the player’s perception of self and their level of disassociation with their character.

If your ultimate goal is to "get male users to act more sensitive and considerate and more people-oriented" I don't think this is the particular path to accomplish this. To effect that you would have to build that into the interaction with the NPC's and game-play where the players are rewarded for acts consistent with what your looking for and penalized for negative behaviors.

14.

"When I'm online, I try not to assume anything. I do make assumptions, often based on behavior rather than appearances"

I really don't see how behaviour based judgements can be considered assumptions? If someone is acting like a 12 year old, they *are* a 12 year old. I don't particularly care if they inhabit a 28 year old avatar in the Real World. Similarly with gender: if you play a female avatar, you ARE female. You may claim to me that you are a male trapped in a female body, or have some other embodiment which is male, but that doesn't affect the fact you *are*, in this medium, female. The fact we know everyone has chosen their gender, if anything, strengthens this relationship.

These views only appear radical if one subscribes to the traditional A/S/L chat room philosophy. In a virtual world, I'd reply 6/Male/Nujel'm.

- Brask Mumei

15.

Brask is hitting on something that takes a small turn for the weird, that is, that we ARE effectively the sex we represent, which digs a big deeper at what sex IS, whether its strictly physique as we most commonly identify it, the combination of physique and behavior, or whether it can be behavior alone - the latter being the most interesting. Are we male and female IRL merely because we represent ourselves as such? If so, wouldn't that apply in VW's as well, given the apparent ease of putting on a different virtual anatomy?

Much like Bart posted early on, I used to play a female avatar in UO (I am male) because it was very much in line with the type of character I wanted to play. I was a griefer well before the title was coined, and my greatest gaming enjoyment came from PK'ing, looting and smack-talk. To my mind, the game was all about adversarialism, and the point of the game was to Win by defeating my opponent, physically, mentally and economically. Playing the role of a friendly female often added to the impact of the whatever chicanery was afoot when all was revealed later and my opponent was dead and stripped. So I guess my anecdote is this: gender bending wasn't a reflection of personal sexual confusion so much as a practical device. I don't perceive my avatar to ever be 'me', but rather as a toy or sorts I'm marching around in the sandbox. I choose females because its easy to manipulate my relationship with others. I can more easily acquire free equipment, accurate answers, and naive young men more easily fall into PK traps. The second reason I choose female avatars is because they are simply more pleasant to look at. I prefer female avatars for the same reason I prefer the company of beautiful women IRL.

Fascinatingly enough, gender-bending for the purpose of manipulating favorable perceptions spills over into the eBay economy as well. In the last year, when UO gold dealer 'Ingotdude' of UODudes.com got banned for exploiting, saw his reputation destroyed for collaboration with Blacksnow's Lee Caldwell, he repackaged himself as 'UOWoman' and resumed business. At least, this is the common understanding in the seller community. Pretty savvy I suppose, but it gets back to the underlying motivation for gender-bending, which I think, more often than not, is just a convenient 'trick' that males with no sexual identity problems use to confuse males who can't conceive doing this themselves.

What I see happening looks more like Males warring with Males, with some socially exploitative Males in drag. Surely this explains the bulk of the behavior, as opposed to sexual identity confusion?

My wife routinely plays EQ and DAoC, and it seems she only creates female avatars that approximate her appearance, so she's from the other school of thought - that avatars are projections of self. She finds me a bit weird that I create females, but I can't say her behavior is stereotypically female either - I heard her mumble "Die you Bastard" as I was writing this. :)

16.

Brask puts my position more succinctly than I. I would add two quick points:

-Gender doesn't map perfectly from RL to VW. What 'female' means in a VW is not the same thing as at your RL office. Gender becomes both a more pure and more fragmented concept.

-This once again highlights an axis of gameplay approach: on one end the Immersivists who become their character and interact with other avatars as a member of that world; and the Proxies, who treat their characters as self-conscious icons or "toys" with which to interact with other humans sitting at keyboards. This is different than the "does/doesn't represent my RL appearance" axis.

It seems to me that obsession over player-avatar differences could only come from the Proxy camp, or at least its discussion in Friedanesque terms. On the Immersive side, it's a non-issue. Though there is undoubtedly some RL psychology to correlate here, we'll need a new vocabulary to profitably study such things among players with increasingly flexible models of self.

17.

Sorry, I'm coming late to this thread and don't have time to read the long postings in detail before posting (unforgiveable, but please forgive if I repeat other's observations)

You might want to talk with Will at There.com. I saw a great set of slides where he demonstrated the view of different avatars of exactly the same event. In "my" world-view everyone is looking at me, even when I'm not talking. In "your" world-view everyone is looking at you.

I seem to recall that they coded this specifically as a result of user dissatisfaction at being ignored most of the time.

18.

On a related theme there is a current thread over at MUD-DEV called ‘Character Perceptions’ which is on the pros and cons of ones perception of a virtual world being based on your characters traits e.g. if you’re a good thief you might be able to ‘see’ rich people more easily – fascinating stuff. Here’s the link.

19.

What a fascinating idea. Perhaps gender-switching without the player's knowledge is too drastic, but what if you made more subtle changes?

I confess to sometimes prejudging other players based on their choices in character customization. If they tend toward extremes in physical characteristics, for instance, I may tend to get an impression that they are likely to be less interesting than a character that exhibits more understated qualities. This will tend to affect how I treat them and thus make it more likely that they will act in line with my perceptions, creating a sort of feedback loop that could perpetuate negativity.

But what if that hulking, strong-jawed, Adonis with bright, spiked, garrish armor appeared as a serious, scholarly, and reasonably dressed warrior to me? Or how much could you shade the avatar toward this sort of thing before someone noticed? How much do you need to shade it before it makes a difference?

Too bad that when the l33t d00d opens his mouth, my impression of him is likely to be based much, much more on what and how he types than on his appearance. Which may signal that throwing in a real-time grammar and spelling checker might go farther toward forming more positive impressions than would changing appearance.

Still, it creeps me out a bit to think about manipulating or being manipulated in such a manner. At its core, it may not really be much different than advertising, but it certainly feels a bit more intrusive to me.

--Phin

20.


Phin > Still, it creeps me out a bit to think about manipulating or being manipulated in such a manner. At its core, it may not really be much different than advertising, but it certainly feels a bit more intrusive to me.

It’s interesting that you object to being manipulated in respect of the visual experience. People in MMOs are being manipulated all the time. Just look at the current threads on Shorts Law or the part of the Raph interview I highlighted in the Raph on the Spot. All of this social engineering is for _better_ VWs, but the point of Character Perception idea was to enhance the experience too.

I wonder why this feels more manipulative, maybe because it feel more personal, altering our perception of the world really is Descartes demon made real yet as you note advertising does a similar thing: a cool brand of jeans look cooler; and as someone once said – how come my slick aerodynamic mobile phone starts to look all boxy and wrong after only a few months.

Ren

The comments to this entry are closed.