As strange as it might seem, VR technology was a natural offshoot of the counterculture - a reliance on small-scale technologies to create a disembodied, connected global consciousness (think LSD, funky musical instruments, Whole Earth catalog of gadgets, and personal computers). As an aside, Fred Turner has a wonderful book coming out on the intersection of the counterculture and personal computing.
Of course, contemporary virtual worlds reveal an insistence on embodiment. Most virtual worlds are avatar-based. Where the counterculture was heading towards the liberation of disembodiment, the opposite has taken hold. In a recent study (PDF) at Stanford, we explored this tension between liberation and embodiment in virtual worlds.
There are many well-known patterns that goven how people interact in the physical world. For example, within a social distance of about 12 feet, the closer two people are, the less likely they will maintain eye contact. This is the elevator effect. Proximity and eye gaze are both signs of intimacy. To keep things in equlibrium, eye gaze compensates for distance when we stand too close to another person. Another well-observed pattern is that men maintain less eye contact with other men than women do with each other.
The question we looked into was: Do these patterns carry over into virtual worlds where people move with mice and keyboard instead of arms and legs, and where virtual gender need not match the real gender of a user? We used a script to collect positional and orientation data on social groups in Second Life over a period of about 2 months.
Our data showed that many patterns of physical interaction in the real world carry over into the virtual world. In other words, our insistence on embodiment in virtual environments structures social interactions in these worlds in ways that we may not consciously be aware of. On the other hand, this implies that virtual worlds may be useful platforms for studying things even as visceral as the rules of physical interaction.
-------------------------------------------------
Yee, N., Bailenson, J.N., Urbanek, M., Chang, F., Merget, D. (in press). The Unbearable Likeness of Being Digital: The Persistence of Nonverbal Social Norms in Online Virtual Environments. The Journal of CyberPsychology and Behavior. PDF
That's really interesting, but I do wonder if the Second Life interface is the best analogy for comparison.
I tend to find the chatline very detached from the avatar, so I spend most of my talking time looking at the bottom left of the screen without much of a thought to where I happened to have parked my av.
I think the 'bubbles' option is probably better, did you make a distinction in your research?
Posted by: always_black | Aug 08, 2006 at 07:49
If you're interested in counterculture and cyberculture, Turner's book is not the first (despite the claim on the linked page) to explore the field- Markoff's What the Dormouse Said is a pretty good collection of anecdotes about the ties between the 60s and the birth of the PC.
Posted by: Luis | Aug 08, 2006 at 10:06
always_black,
I just skimmed through the paper (nice work, Nick) and while I agree that focusing on the chat COULD cause a user to potentially ignore nonverbal cues like IPD and gaze, this should affect all interactions rather equally. The study here showed that such cues did seem to extend to a virtual setting in a predictable way.
This has got me wondering whether different UI designs will affect any of this. Will using a chat bubble result in a larger IPD so players can have an unobstructed view of speaker and bubble? In games like EQ2, your gaze tracks the person you have targeted (to an extent- no exorcist-like neck behaviors... usually). Does that "default" behavior cause people to stay farther apart, and does close proximity cause a person to actively target away (break eye contact).
I also find it interesting that IPD appears to be identified through avatar location, and not camera position (which would have been difficult to determine in the study.) If that's true, then the association of "self" with the avatar is rather strong. I would have suspected that such a proximity-based metric would be distorted as players put use the camera location as their spatial reference, not the avatar's.
Posted by: Chas | Aug 08, 2006 at 10:23
"this should affect all interactions rather equally" - all things being equal, but I'm cursed with an inability to type without staring at my fingers whereas some just know where the keys are.
I wondered about the detachable camera too, it's one of my bad habits to go astral-projection while I'm supposed to be sat attentively listening.
Also, I might be mistaken here, but in SL doesn't your avatar automatically look at the last speaker when it's idling?
All in all, I wonder if there isn't a bunch of work to do analysing the interaction between people and the program's interface before you can determine whether such interaction and real interpersonal interaction are comparable.
Posted by: always_black | Aug 08, 2006 at 10:43
Actually, wouldn't it make more sense to compare human to avatar contact, rather than avatar to avatar?
Posted by: always_black | Aug 08, 2006 at 10:47
Posted by: Mind Booster Noori | Aug 08, 2006 at 10:50
In SL, *you* (the player) aren't ever really looking where your avatar is, unless you have gone into 1st person mode which, I understand from several studies and "askin' around," very few people do. My avatar may look like he/she's looking at you, the floor, somebody else... but my camera is pointing at... anything else. I (my avatar) can be sitting at the bar, staring at my (avatar's) drink, but my camera can be churning around like an invisible Tinkerbell, checking out the crowd, going upskirt, examining tattos, wandering around outside... whatever, wherever. This is the "visual" equivalent of the "my avatar is here in front of you, but I'm actually IMing with nine other people you can't see."
Which is pretty much a departure from creating an "embodied" virtual reality. Question for the peanut gallery: does the fact that you have an avatar mean you are strictly "embodied," if that body isn't limited to actions/interactions that are fairly linked to what a body could do in RL? If my avatar in SL is sitting at the bottom of a small lake, but my "invisible eyeball" is caroming around a nearby disco, and I am chatting with four people simultaneously, none of whom are anywhere near my avatar, or my wandering eye... am I in any way truly "embodied?"
I think not.
Posted by: Andy Havens | Aug 08, 2006 at 12:26
As far as controlling levels of intimacy, I agree with Noori. In places like SL, focusing one's emotional energy is a work in progress.
But I think it's a more complex affair as far as embodiment is concerned. Just because we happen to be represented by an avatar in MMOGs does not necessarily lead to embodiment, "the fusion of substance and spirit," in a meaningful way. In fact, I think that experience with avatars helps to radicalize the essence of disembodiment, being "divest(ed) of material existence or substance."
Posted by: Kliger | Aug 08, 2006 at 12:35
Andy
I think avatars can 'embody' in a certain sense. It's often a product of creative choices you make regarding appearance, demeanour etc, so it's a kind of personal symbol. But I think it's an entirely different beast to your um, actual body and you use them consciously and unconsciously in entirely different ways.
Posted by: always_black | Aug 08, 2006 at 15:43
Andy has hit the bed of nails squarely in the center. Experienced virtual world users exploit the technology for new approaches to interaction, especially if chat is moving at typing speeds (boring to everyone except the typist).
My belief that as users gain experience they also tend to exploit the disembodiment potential of virtual worlds leads to some interesting speculation about why people prefer 3rd person to 1st person views - above and beyond the obvious arguments.
I don't believe we can begin to study Nick's original proposition until we have a 1st-person-only platform that uses VOIP rather than typing as the medium of communication. Fairly high fidelity facial features with "facial emote" controls wouldn't hurt either.
Posted by: Arnold Hendrick | Aug 08, 2006 at 16:06
The "human to avatar contact" that always_black mentioned and the 1st-person-only platform that Arnold mentioned exists in the form of immersive VR. And studies in IDP have also been done in immersive VR, yielding similar results (see PDF)
I think human<->avatar and avatar<->avatar are equally interesting questions. I think the interesting thing the SL study shows is that even in low-fidelity virtual environments, physical rules of engagement still apply to a certain degree.
Posted by: Nick Yee | Aug 08, 2006 at 19:22
Fascinating study! I'm especially interested in any "gender differences" based on avatar gender. It supports the sociological 'doing gender' perspective over biological perspectives (assuming your results aren't accounted for by the physical gender of the players).
Did you measure the "yaw" of the avatars' bodies or heads? (I'm assuming bodies, but your use of the term "mutual gaze" and Figure 1 suggest heads.) From my qualitative analyses of screen-capture video, I'd agree that players use "body orientation" in virtual worlds much like they do in the physical world. However, I find that the use of "eye gaze" (or "head orientation") differs considerably from the physical world and is heavily impacted by the particular avatar system. For example, in Second Life, avatars' eyes follow the mouse pointer. They'll "look at" another avatar if you place your pointer over it, but as soon as you move the pointer up to a pull-down menu, down to the button bars, or over to an open window, it looks at those user interface objects instead, regardless of what is happening in the conversation. And of course the other player can't see any of these objects. Consequently, I find it very difficult to make my avatar maintain eye contact with others in Second Life. In contrast, a very different system is EverQuest II (or Star Wars Galaxies) in which targeting another avatar locks your avatar's "gaze" on its face. Here it's too easy to maintain eye contact so avatars end up "staring" at each other to some extent.
"Hall (1959) estimated that social distance extends up to 12 feet (or 3.66 meters). For the following analyses, we used a social distance of 3.66 meters as the cut-off for whether 2 avatars are considered to be a dyad" (p. 10). Here aren't you assuming one of the very things you're trying to test? Or did you explore other cut-off distances as well? I wouldn't be surprised if this distance was greater in Second Life and other MMOs than in the physical world. Furthermore, interpersonal distance in real life is influenced by the noise level in the setting (we stand closer to each other at parties and concerts so we can hear each other speak). In MMOs might it not be influenced by chat distance (the distance within which I can see other players' spatial chat messages)? What's the chat distance in Second Life?
I applaud the study for tackling a set of hard problems: Can conversations among characters in MMOs be detected automatically? Is detecting characters standing close together, facing each other and saying something good enough as a proxy? How do you detect multiparty conversations? Is linguistic analysis of chat messages necessary? There's still much work to be done...
Posted by: Bob Moore | Aug 08, 2006 at 22:56
Here's another question related to embodiment; anthropomorphism in games, game control objects, avies, etc. How much do we tend to "personify" and ensoul those objects we use to play games with? If I'm any measure... I'm guessing "lots."
I didn't mean to imply, always_black, that the emobdiment of an avatar had to come anywhere near close to a physical embodiment to "count" as such. Just that it often doesn't, even when there's a picture of "a body;" i.e., having an avatar body doesn't necessarily count as a locked, embodied game viewpoint. Or, if you look at it this way, I have one "magic circle" when I play SL, and my avatar has a different one. Oooh.... freaky.
Back to anthropomorphism, though.
When we feel "embodied" in an object that is not us, we are self-anthropomorphising it. We may not be emodying it with the traits we ourselves exactly have (Quick! Rush to the "Horde is Evil" post!), but we are putting *something* of ourselves into whatever "thing" it is that is being used to play the game, eh?
Language supports this. We don't say, "Place another checker upon mine, such that it is now a king." We say, "King me." Even the checker, a simple disk of plastic is, to a very slight degree, an anthropomorphised avatar of our game-will. How much more so any object/sofware-bot-thingy that actually resembles a person?
Posted by: Andy Havens | Aug 09, 2006 at 13:12
Although it's nice to imagine the birth of VR as being closely tied to the counter-culture, that's not quite accurate. While there were some close ties in the late-80's/early-90's e.g. Jaron Lanier et al, the origins of modern virtual reality go back much further and are related more to the military-industrial-education complex, as is computer graphics in general (think Ivan Sutherland, Thomas Furness, and Frederick Brooks). Refer to Howard Rheingold's "Virtual Reality" for a thorough history of the early days.
Posted by: Christian Greuel | Aug 10, 2006 at 13:15
I have always felt that having an avatar in SL was detrimental somehow--haven't thought of a better way to do it yet, but still. You have no fine control over your body--only set patterns that you have to program beforehand. You don't even have control over your gaze and your chatting at the same time. I feel like yes, the avatar adds to a "realistic feel" for the game, but perhaps that's thinking about it the wrong way, just as thinking about electric lights as lamps without fire wasn't the most efficient way to light a room. Maybe if we transcend the avatar, we can truly create a new communication channel.
Posted by: Christina | Aug 13, 2006 at 15:26
Andy, the avatar is still definitely embodied. Or, he is an ensouled body. He is the perceptive soul (http://www.sacred-texts.com/cla/plotenn/enn391.htm)
That is, his body and perception-taking-in-apparatus is all part of a kind of cloud, if you will, which constitutes his avataric body inworld.
His powers of flying, alt-zooming, taking in 9 IMs at once, receiving inventory, etc. etc. are all still part of a "body," just not a traditional, meat-world body in the modern sense, but a body/soul entity in an ancient sense in many religious teachings.
As long as a meat-world body's limits of eyeball and ear perceptions are a factor, the avatar will remain a body, though a higher-powered one.
Other things might be added to his powers. For exampel, what if I had control of a bot, that could go out and scan for land that I'm interested in buying, and keep a running ticker on my HUD view of what's out there so I can be ready to pounce with p2p (In fact, some land-swoopers in SL apparently have scanners and stuff working almost like this, or they just obsessively refresh the land for sale list lol).
Or what if I have an interesting group event where I can't attend because I'm busy elsewhere, but some prim or something scans the conversation for me and delivers it, or key words, on some bubble or blow-uppable thingie on my user interface, or maybe it aggregates blogs for me or whatever, it's still part of my "body" however because I told it to scan for this land or that blog, not the other one.
Then I suppose it gets interesting. What if I opt to get the lifelog and myware 'print out' if you will from somebody else similarly all plugged in -- and in trying to purchase that land, I access the data that the rest of the people in the sim have in their repetoire. Of course, now I'm entering that hive mind and smart mob so beloved by tekkies. And when I do that, I get the chat bubbles/thingies saying "OMGZORZ don't buy on this sim there's a club with events holding 40 avatars hiding in that building that looks likes a shack on Tuesday nights!!!!". Of course, these same trackable zoomable things can invade privacy and ruin reputations with false information and spread libel, too, which is why I remain critical.
Christina, I think we already transcended the avatar. He's just a base. You zoom out or zoom in from there. They can put in more puppetry and they seem to be drilling deep on putting more fine-tuning to puppets at SL with their new stuff they showed off at SIGGRAPH. Or you could go more shallow and just make world-visiting a giant browser experience where my desktop and browser is all set up and I never dangle an avatar anywhere, I'm just in my chair in RL like in a spaceship. I see some REAL downsides to that, however.
It's very visible to me right now in the way in which many new waves of people are coming into SL on the free accounts -- that means younger, and poorer. Many of them treat the all of Second Life as a kind of giant video game backdrop, all the people they meet in it as NPCs who are expendable or beatable, and their own experience as a kind of solo pilot playing a solo video game. Hence, their main mode of interaction is to ride around, shoot, swear, crash, ruin stuff, try to goad people to see if they'll talk, etc. It's made life pretty unbearable in places, unleashing a huge backlash of bannings based on verified/unverified account status, which the Lindens handily put into the land menus now, and red ban lines everywhere.
Bob More, here's a discussion about Nick Yee's study and the reaction of myself and another rentals agent in SL, since we're able to observe lots of people:
http://blog.secondlife.com/2006/08/04/this-just-in/#comments
Basically, I do see gender differences in how the ensouled/body avatar comes into spaces and territorializes them or famliarizes with them -- males and would-be griefers have more walking, flying around and sharp darting motions; householders/homesteaders/females setting up house sit more and stay in one place more, etc. In fact, I have developed a kind of radar I guess, as probably many have unconsciously, where you can tell the people who are males using the female avatars versus the real females, though of course this isn't foolproof -- and more importantly, you can spot trouble coming from people likely to grief an event.
There are also "gameplay" cues that come into play, for example, if I'm flying around my plaza and I see a kid with a gun suddenly rez a cube and sit on it, I figure he's adopting the crouch pose that war-gamers and shooters adapt to prevent themselves from being "orbited," or sent into orbit away from the parcel.
In fact, the whole reason that SL can be the place where "the men are men...and so are a lot of the women" is because adapting the female soul/body patterns just seemed a natural for some, being in this world, and also I would submit, their chosing of the opposite gender does probably speak to aspects of gender that are simply not known or studied yet and will be some day.
Posted by: Prokofy Neva | Aug 13, 2006 at 17:58
@Prokofy: If you want to define an avatar's "body" as "The sum of all things that a user can do when in a VW, and where the term 'avatar' is applied in any way to the manifestation of the user's will," then, of course, everything you do, whether visually, aurally or any-other-wise connected or disconnected from your avatar is synonymous with your embodiment of the avatar and his/her/its own embodiment of self. That's just another way of saying that "I am my avatar, and he/she/it is me and I embody him/her/it and we are all together."
The point, I thought, was something slightly subtler.
For example. If you and I are both playing SL, and our avies are in a shared visual space, the same room, let's say, we are sharing one aspect of our embodiment, correct? I can see your/your avie's visual representation of self. You can see mine. When I gesture, you can interpret the gesture. If we care to do something visual together -- dance, jump, fight, etc. -- we share a visual experience based on how we have chosen to embody our experience. If, for example, we have compatible gestures, programmed dance moves, whatever... we can do some visually neat stuff. If we want to chase each other around and be foolish, we can. It's all part of our shared visual space.
Same holds true for open chat and sound (to a certain degree in SL, but let's imagine that we both have compatible sound hardware). I type a message, you read it as I have typed it. I play a sound, you hear it. We share the space that our avatars inhabit in their embodiment of our will and their being.
If, however, I begin to do something -- anything -- with the interface of the VW, the UI, that does not allow for your interpretation or interaction via your embodiment of the shared space, I have begun a "disembodied" process. If I am standing (dancing, whatever) in the same room with your avie, sharing that bodily space, but use the UI to IM with a bunch of other people, and pull my view back to watch what's going on in another room, and -- as you suggest -- have monitor bots and pop-up windows and things going on... those are all events that are related to me as the player/user, but not (necessarily) my avatar.
Now, of course, I can always go AFK and get some coffee, talk to my parrot, tuck in my son, etc. etc. in RL and those are clearly "disembodied" events from an in-world standpoint... But I would argue that when you are doing "SL stuff" that those around you are completely unable to even recognize, let alone participate in... you have become, from them, disembodied to some extent. Maybe 100% in some cases.
Now, this holds true in RL as well, of course, but only inasmuch as I can carry on a phone conversation or multitask or Blackberry while in the presence of others. And in many circumstances, this is considered quite rude. In some cases it's not, sure. But if you and I are having a nice chat, or dancing, and I pick up my phone and start calling someone else to have another casual conversation... nice! But that happens all the time in SL. Why/how? Because the avatar is more easily disembodied by the technology. You have no idea that the outside IM is going on -- except for the fact that the avie just sits there like an idiot for minutes at a time.
Hell. There are times I feel disembodied in RL when I'm trying to do just one thing at my desk at work...
Anyways... I don't mean to imply that the avie doesn't have his/her/its own body. Just that not every task associated with life/play in VWs/MMOs is associated with or can be linked to the avie. Which, to me, is the essence of embodiment.
Posted by: Andy Havens | Aug 13, 2006 at 21:52
Is there a Turing test for parrots?
Posted by: always_black | Aug 14, 2006 at 05:38
@always_black: Yes.
Pirate: Polly want a cracker?
Parrot: Look out behind ye, Capn'!
[Captain ducks, avoids being beheaded by scurvy dog, spins and discharges pistol into same.]
Pirate: Now that's a true friend y'are, parrot! Truer than most (kicks scurvy dog).
Posted by: Andy Havens | Aug 14, 2006 at 09:30
Re: "If, however, I begin to do something -- anything -- with the interface of the VW, the UI, that does not allow for your interpretation or interaction via your embodiment of the shared space, I have begun a "disembodied" process"
This strikes me as an artificial separation of the sorts of things you do in a game as an avatar, from the sorts of things you do in RL as a human -- they are analogous.
In RL, I might be sitting at my desk, and I might begin to compose a memo in my mind. I've seemingly "left the shared interactive space" of my RL office where I talked to and gestured to others, and entered the "user panel" of my mind where I mull over pictures, words, concepts. You can't see what I'm thinking if you are at the next desk over, though you might put together clues. But...I haven't left my body then, I've merely focused inside my mind while I compose something on my internal user panel.
I remember Will Wright would talk about playing the sims (the avatars) as our "investing a toy with consciousness." We would begin to say "I" about that sim, and feel that things that happened to him happened to us, too.
You seem to be trying to stake out a territory for where the avatar and his collection of actions and thoughts from his body are inhabited, and a territory for where the human and his collection of actions and thoughts from his body are inhabited as some kind of real dividing line -- but they are blurred at best.
No doubt there is a cloud between the two or a grey area as people phase in and out.
You seem to want to sharply delineate the avatar's body as only that which manifests itself to others in the shared space; yet you wouldn't so sharply delineate the human's thoughts and composing and computation internally in the same way.
I'm simply identifying the avatar body differently, drawing the line further out. While I'm on the game user panel, reading the stats of my skills or checking the land prices or having IMs, I didn't leave my avatar body to go to my human body, so to speak, I'm actually in my avatar's mind/soul, not his body. Everything I'm doing is related to the avatar's second life, not my first life.
While playing that game or inhabiting/exploring that world, I'm in that avatar body/soul. Only when I got up and went AFK or logged off would I leave his body/soul.
And if there were no game, just because I go on the telephone with someone in RL versus talking to them in the room doesn't mean I "left my body". Nor would I have left it if I merely read a newspaper or sat lost in thought.
So I haven't left the avatar's body merely because I retreated into the user-panel space and experience.
To be sure, yes, the avatar's embodiment is a fragile concept, constantly torn asunder by the typist's need to go on the phone "in real" or go to the bathroom, leaving the avatar idling with AFK balloons over his head. Still, fragile though it is, it is an embodiment, and while engrossed in that game, the human sustains it.
Posted by: Prokofy Neva | Aug 14, 2006 at 13:13
@Prokofy: Interesting. I hadn't quite thought of it that way. That anything done in the "inside" of the VW/game counts as "inside the avatar's head." And, as you say, what's in your head is in your body in RL, 'cause your head is, for sure, part o' your body.
But there is still a blurry distinction, even in RL. If you approach me and I am so lost in thought that I don't sense your approach, you might wave a hand in front of my face and say something like, "Hello? Anybody home?" And if I'm asleep, or drugged out, or unconscious or concentrating on something that requires use of various senses (driving, let's say), you might have a similar experience in RL to the one we're talking about in SL.
At which point I have to make the point that if there is no disembodiment possible in RL, then of course none is possible in SL. If I can't "get out of my head" in my kitchen, then my avatar can't get out of his/her/its head in SL.
To me, though, the concept of disembodiment is the idea that, to some degree or another, I *have* made a distinction or separation between mind and body. After all... what else would be "disem-ed" from my "body" except my mind? There must be physical, neural connections, of course... or else I'd die. But the concept, as I understand it, is to somehow "project" my consciousness, to some degree or another, into a realm not inhabited by my physical self, nor shared equally (necessarily) by others in the presence of my physical self.
If that's the case -- if there is a legitimate definition of disembodiment in RL -- than I'll just ask the question: is there a virtual/SL equivalent? Or are we, by process of using a "mind tool" such as a computer, as disembodied as we can get to begin with, and, therefore, any distinction between having an avatar, not having one, invisible "space ship HUDs," IM clouds, bots, UIs, etc. is moot from the point of view of "embodiment?" In that case, it's all about my fleshy brain in my fleshy head touching my sweaty mouse on my woody desk in my muggy room. And all else is... other.
Posted by: Andy Havens | Aug 14, 2006 at 17:28
One of the more disconcerting features of There.com is the fact that avatars stand there and nod as they "listen" to someone speaking. At first this behavior feels quite natural and very reassuring when you are the speaker, but when you notice your avatar nodding agreement to what someone else is saying when you have no intention to do such a thing, you suddenly develop a strong disconnect between self and avatar. I agree with Arnold Hendrick that more advanced players will exploit disembodiment; still others, I think, will master puppetry (John Malkovitch, anyone?). But if the platform acts out these behaviors for you, it robs long-term skills-building in exchange for short term (deceptive) immersion.
Posted by: Gene Koo | Aug 17, 2006 at 23:24
For me, the interesting bits of this paper were two: one, like Prokofy Neva pointed out, was that there seemed to be no difference between the behaviour of male/female avatars (no matter what physical gender is really behind it!) and of male/female human beings. Unless, of course, in this research all the real life identities of the participants were, in fact, known in advance.
If not, this is another step into an emerging hypothesis being forwarded that our human behaviour is also determined by visual clues. We all have heard about "monkeys imitating humans", and picking basically things from us, just by observing us. It's also clear that most children pick up behaviour patterns from their parents and later from partners at school. And finally, we also know that people going to live in another country will over time pick up patterns from the country they live in. The ability we humans have to "blend in" is rather one of our characteristics.
When coming into a virtual world, however, things become more complex — and also interesting at the same time. First, we have no constraints on our physical appearance — an important consideration for, say, the furry community, which finally can act and behave like their animal of choice, and not as a "human clad in an animal suit". Besides the obvious appearance, or the gender-bending issues, there are obviously many others — changing sexual orientation, age, etc. sometimes connected to what you wear (or refuse to wear), and, in some degrees, depending on the particular community you join (if you join an Orc community keen on role-playing, you don't expect to discuss Kant and Descartes with your fellows).
But then there is the reverse issue. Why do we sit down if we want to talk during a long time? Why we knock at doors before entering? Why do we ask permission (in the case of virtual worlds that allow it) for people to be teleported to your location? Why might we fly around (typical of Second Life) but change to walking when you're getting near to a destination where you will interact with other people? (Think on how many social norms have evolved for us to chat and interact while hovering 100 metres above the ground! You guessed it — none :) ) It's not that we don't, effectively, do all these things — we do even more, getting all our habits and ticks into virtual worlds — which is quite clear that we do. The question (left unanswered) is "why?"
On the real world, of course, we might be constrained by a specific set of norms and rules of our community, and the answer to the "why" is "because otherwise I would ostracised by the community I am in" (ie. being seen as rude, unpolite, or even a sociopath). But the "why" is not so easily answered on virtual spaces. Usually, the best we can answer is "why not?", meaning precisely that, "if we behave in a certain way in the real world, why should we act differently on virtual worlds, since we're the 'same' human beings after all?" (to which a skeptic would obviously counter-answer "because we can!")
The second question that popped up after a most interesting in-world discussion of this paper is if Second Life's creators, Linden Lab, were actually aware of interpersonal distance factors when creating it. In a very strange — but familiar! — way, the camera seems to be positioned "just right" to facilitate interaction. Sure, you can move around — and most people will do that! — but you'll present yourselves to others as someone "just at the right position" to the person you're talking with, facing them, as you would in real life — and this paper seems to confirm exactly that. Put into other words: while *your* camera be wildly swinging around, your partner in a conversation will see your avatar focused on you, keeping eye contact, and the apropriate facing and distance for proper conversation. And the interface actually *encourages* just that.
This might seem "obvious" — but it isn't. Other virtual worlds are simply not built that way. You might get camera controls and a way to come "near" the other person — but depending on the interface developer, this might be easier, hard, or impossible. When trying out other 3D graphical platforms, this is something that utterly confuses me, especially when being "tutored" by a veteran user — I'm a newbie, I don't want to appear rude, but somehow the interface doesn't allow me to be placed in a "proper" position: facing them, being at the appropriate position. These interfaces look to me as particularly "clumsy" — they minght be appropriate for some other uses, but they are worthless as enabling and facilitating avatar-to-avatar communication.
These feel like if you somehow had a weight of 10 tons on your head and you had to try to balance it (or be smashed by it) while engaging in conversation with someone else. It simply is too cumbersome. Now, is Second Life's interface design deliberately done to facilitate "natural" conversation — or are we just lucky? :)
Posted by: Gwyneth Llewelyn | Sep 10, 2006 at 14:39
The University of Chicago Press website has two excerpts from From Counterculture to Cyberculture: Stewart Brand, the Whole Earth Network, and the Rise of Digital Utopianism by Fred Turner. You can read the introduction and an excerpt about the Whole Earth Catalog and the emerging digital culture.
Posted by: Dean Blobaum | Sep 12, 2006 at 15:36
great info! keep on blogging man!
Posted by: Lea Sterling | Oct 16, 2006 at 12:20
Well, just took three months for this report to make the mainstream:
http://www.nytimes.com/2006/11/16/fashion/16space.html?em&ex=1163912400&en=9005fdad1a8552d4&ei=5087%0A
Currently the top story in the Times-- good on you, Yee!
Posted by: Hamlet Au | Nov 17, 2006 at 03:34
I have only just read about Yee's research in a recent story by the Associated Press (and subsequently picked up by the CBC) due (it would seem) to its appearance in the Feb 2007 issue of CyberPsychology & Behavior. I have written about it here: http://fpsl.wordpress.com/2007/02/24/cbc-article-reports-on-flawed-research/
Posted by: Fiend Ludwig | Feb 24, 2007 at 14:47