From: vanevery@blarg.net (Brandon J. Van Every) Subject: PHIL: measuring "immersion" in 3d and text Date: 6 Apr 1996 23:57:55 GMT Message-ID: <4k70e3$7vk@guysmiley.blarg.net> Organization: Blarg! Online Services 206/441-9109 SUMMARY ------- The possibility of measuring "immersion" in both 3d and textual virtual worlds is discussed, relative to the following definitions. A "virtual world" 1) must utilize computer hardware 2) must map a spatial domain 3) must allow the spatial domain to be accessed non-sequentially In addition, "Virtual Reality" 4) must allow the user to traverse the space in real time Additional input is sought on how to further refine this definition of "Virtual Reality." BODY ---- Please ignore my previous response to Marc on this ongoing subject. It was supposed to be a private e-mail and got snafu'd. A lot of the debate about whether "textual MUDs are VR," and indeed whether anything and everything is VR, centers on the vagueness of the term "immersion." Even if it has an exact definition in some circles, it's an overloaded word. When are we speaking of "visual immersion," when are we speaking of "psychological immersion," and how do we know that the two don't overlap greatly? How much of immersion is based upon the biology of vision, and how much of it is based upon the constructs of language and culture? How essential is 3d? How much is based upon the cognitive maps that we construct about "space," such as when someone _tells_ us about a space, instead of showing it to us? I ask "how much" because I think it's possible to measure this quantitatively. Asking 10,000 people what they think "immersion" is, is not a bad idea in and of itself. However, the analytical criteria must be properly devised, and the biases of the pollsters must be taken into account. Automatically assuming that "VR must have 3d" is not a scientific response. It is a definitional response, or a political response. We should be looking for ways to _prove_ that VR "must have 3d," or at least that it is _the_ overwhelming factor with regards to immersive experience. And if it turns out that other factors are more strongly at play, then we should welcome the analytical insight. In talking with Marc, I came up with the following minimum definition of a "virtual world." My goal was/is to knock out as many "non-VR" things as possible, using the most minimal set of constraints possible. This isn't a definition of VR yet, but it might get there eventually. 1) must utilize computer hardware 2) must map a spatial domain 3) must allow the spatial domain to be accessed non-sequentially 1) knocks out dreams, LSD, and people who want to say that "VR is everything." 2) is essential to both virtual worlds and VR, IMHO. Every single psychological feature of "immersion" that I can point to, deals with people believing that they are within a "space" of some kind. It could be a continuous space, a discrete approximation of a continuous space, and/or a space whose proportionalities are locally warped, but it is nevertheless a space. Our ability to cognitively map spaces has been evolving, both biologically and culturally, for millions of years. We've crossed mountains, planted fields, built houses with familial partitions, architected town squares, and shaped our bodies with ornaments. Space is a major portion of how we construct reality, and so it's foolish to talk about Virtual Reality without talking about space. 3) knocks out books, films, radio, and other sequential media. It does not knock out a hypertext. The nodes of a hypertext _could_ serve as a discrete approximation to a 3d space, if the text itself was written to be such. Also, 3) allows for virtual worlds that have a discrete rather than continuous sampling of available pathways to walk on. If generally speaking you can go somewhere other than backwards and forwards, then it's a virtual world. That, as far as I can see at the present, does it for defining a virtual world. (Although criticisms are certainly welcome.) I've deliberately excluded multi-user participation and real-time interaction. In my view, neither is necessary. Now it's time to take a step closer to "Virtual Reality." VR is more than a virutal world. However, I'm fuzzy about how much more than a virutal world a VR really is. I can agree on at least one criteria, though: 4) must allow the user to traverse the space in real time. But this doesn't really say much more than 1), 2), 3) above. It merely says that the application has to be fast. It doesn't even say that the app has to be interactive. After all, a walkthrough doesn't have to be interactive. You can "walk through," have absolutely nothing react to your presence, do absolutely nothing to the environment, and VR research scientists will still call it "VR." One could try to specify VR in terms of the input/output hardware utilized. In many people's view that's what VR is, after all. One could try to work out a "3d clause." At this point I'm going to let other people try to make the attempt, though. I'm mainly interested if such restrictions will hold up under close scrutiny, and whether they provide a basis for evaluating "immersion," or whether they are just a research preference. Cheers, -- Brandon J. Van Every | Check out Free3d, my 100% efficient, 100% portable | 3d lib, at . 3d Computer Graphics | C++ UNIX X11 Motif | E-mail: vanevery@blarg.net