The Mimesis of Depth

I was reading a chapter in a collection of academic papers about the computer-generated 3D models, and my brain went on a tangent (as it's prone to do) about the value we seem to automatically attribute to the quality of mimesis.

Specific to 3D modeling, it's generally accepted that the computer-generated model of a 3D environment that is depicted on a "flat" monitor screen is a very poor imitation of a real-life environment., and the author's point was that much improvement (and special equipment) is needed before we can realistically depict an immersive environment to the user.

What struck me is the degree to which the user is able to accept a "fake" 3D environment, and I'd suggest it's not because we are applying the perceptual skills learned in the real world, but are transferring the cognitive interpretations learned from similar two-dimensional media: paintings, photographs, and television screens.

The artist learns tricks to "fool" the mind into accepting a flat representation as a three-dimensional one: an object that is smaller is further away, as is a one that lies further along diagonal lines that converges at a horizon. A closer object throws a more distinctive shadow, and an object further away isn't as clearly discernable. If an object overlaps another, it is nearer to us.

We learn these visual cues of "closeness" and "farness" specifically to the two-dimensional representation of depth - such that when we see a highly mimetic painting of an apple, we remark that we could almost reach into the painting and pick it up. I posit that the "almost" is because, in spite of all the visual trickery, there is still a quality that is missing that cues us that what we are viewing is, in fact, a clever forgery of depth rather than actual depth.

So, in the end, the question with which I'm grappling is whether, when we see a "3d" image on a flat screen, we are recognizing it as being three-dimensional from our experiences in the physical world, or because we recognize the conventions that are used to depict depth in a two-dimensional medium.

I tend to think that it's the latter, which may explain the reason why, in spite of all the advances in 3D presentation technology, we find ourselves unsatisfied with even the most detailed and attentive imitations. No matter the degree of effort the presenter undertakes to "fool" us, we recognize the tricks, and cannot accept the mimicry, though it might delight us, for a time, to engage in the willing suspension of disbelief.

An analogy comes to mind - the distortions to the sense of taste. In this culture, and at this time, it is probably reasonable to assume that more people have tasted artificially lime-flavored candy than have tasted actual limes. So a person who recognizes the lime flavor of candy is borrowing upon his sense memory of lime-flavored candy in the past, not from his memory of ever having tasted an actual lime. In fact, if you gave him a piece candy that tasted like a real lime, he might not be able to accurately identify the flavor. Such is the problem with mimesis in the virtual medium.

I generally accept, albeit with some degree of reservations, that a good approach to design in the online media is to mimic "real world" phenomena. The reservations I have is that the online medium has significantly different capabilities and qualities than the real-world medium.

An early example I recall was a video player that attempted to use an interface that was similar to a television remote control, to enable to user to "click" through video presentations in the same way that they surfed channels. This was well before a on-screen programming guides that allowed users to scroll through lists of programs by name, and even view descriptions before making a selection.

In that instance, mimesis has worked in reverse: the television medium, which was presumed to be more familiar to users, has actually adopted navigation controls from the online medium, rather than the other way around - for no other reason than it was easier to use, and just made sense.

It also occurs to me that I have never seen a cell phone interface with a rotary dial - which was at one time the most familiar paradigm for operating a telephone. And the reason is that a rotary dial, while necessary to a physical telephone until about twenty years ago, simply didn't make sense. Like the analogy to artificial "lime" flavoring, a person presented with a rotary dial would have no idea what it was, or how to use it.

And so after that long meander, I return to the notion of creating realistic 3D environments for the Web. Is it really more convenient? Does it really make sense? For the longest time, the answer seemed a self-evident "yes" ... but the more I think about it, the less plausible it seems.