Friday, June 01, 2007

The Clarity, or Not, of Visual Experience

Most people (not everyone!) will say there is some experiential difference between the center of their visual field and the periphery. The center is clear, precise, sharply detailed -- something like that -- and the periphery is hazy, imprecise, lacking detail.

If you agree with this (and if you don't, I'd be interested to hear), I want you to think about the following question: How large is that center of clarity? If you're comfortable with degrees of arc, you might think of it in those terms. Otherwise, think about, say, how much of your desktop you can see in precise detail in a single moment. Consider also, how stable the region of clarity is, approximately how much shifting there is of things from the clear center to the unclear periphery and vice versa. Is it a constant flux, say, or pretty stable over stretches of several seconds?

Humor me, if you will, and formulate in your mind an explicit answer to these questions before reading on.

Dan Dennett suggests the following experiment. Randomly take a card from a deck of playing cards and hold it at arm's length off to one side, just beyond your field of view. Holding your gaze fixed on a single point in front of you, slowly rotate the card toward the center of your field of view (keeping it at arm's length). How close to the center do you have to bring the card before you can determine its suit, its color, its value?

Most people are surprised at the results of this little experiment (so Dennett reports, and so I've found, too). You have to bring it really close! Go try it! If a playing card isn't handy, try a book cover with a picture on it. I've also posted a playing card here, if that might help (image from here).

In doing this exercise, you're doing something pretty unusual (unless you've been a subject in a lot of vision science experiments!) -- you've been attending to, or thinking about, your experience of parts of your visual field not quite at the center of fixation. It's a little tricky, but you can try doing this as your eyes move around more naturally. For example, you might decide to attend to your visual experience of one particular object (maybe the top left part of the banner at the top of your screen, or the Jack to the right), allowing your eyes to move around so that you're looking all around it but never directly at it. How well do you see it?

So here's the question: Has your opinion about your visual experience changed as a result of this little exercise? And if so, how?

I have a little bit of a wager, you might say, with a colleague of mine about this.

11 comments:

Justin Tiwald said...

Huh. I would have guessed that my center of clarity would be a circle of about 5-6 inches in diameter. In fact it's less than one inch.

In light of the experiment, I think the experiences I once would have described as "seeing the card clearly" are really better described as a combination of the following three things:

1. Seeing a tiny portion of the card clearly.
2. Remembering how the other portions looked when I saw them clearly a moment ago.
3. Having that memory confirmed by (or at least not contradicted by) my fuzzier peripheral vision.

I had a prior hunch that "seeing" involves a certain amount of reconstruction. I think the experiment really confirmed this, and highlighted how extensive the reconstruction is.

Hope I made you a few bucks!

Quirinius_Quine said...

I can't say I'm surprised, because I already read about the experiment in Consciouness Explained some years back. I have a question, though: Has there been any work done on the constancy of the size of the 'clear region' under different conditions? For example, perhaps the clear region is small during this experiment because your eyes are stationary while you move the card. The region might exapand when you're shifting your gaze around frequently, perhaps because your brain is taking in more information about that area and can integrate it into a larger region of high detail.

Daryl said...

Hi Eric. I see no reason to assume that the center of clarity, if there is such a thing, has any dimension whatsoever, or that vision comprises discrete units of visual space. Vision and visual space are as distinct from each other as movement and distance covered. Think of Zeno's paradox.

Pete Mandik said...

"Has your opinion about your visual experience changed as a result of this little exercise? And if so, how"

When I was first exposed, years ago, to Dennett's card trick, I found it surprising and it did indeed change my opinion. I would have thought that determinate colors, if not shapes, made it into the periphery. If I'm looking at a uniformly colored stripe that horizontally extens all the way accross my visual field, I would have previously guessed that it would be determinately colored throughout. I can't give any kind of serious size estimate, though.

Daryl said...

An interesting variation of this experiment, I think, would be to use two cards coming from opposite directions at once. Another variation would be to fix your eyes on a symbol that changes randomly, for example, a number or letter, while you move the card closer to the central point, and determine whether you can identify the card and the changing symbols at the same time; whether, in doing so, the space of clarity has the same arc as when you fix your eyes on something stable; and whether the minimally perceptible size of the symbols would differ from what you would normally be able to perceive under normal conditions.

Eric Schwitzgebel said...

Thanks for all the comments, folks!

I'll refrain from expressing which direction my bet was in for a few days, in case some other responses come in. But quirinus and daryl, I agree those would be interesting experiments to do. In fact, some of them have been done -- with respect to our visual capacities (as opposed to our visual experience). For example, Mack and Rock and colleagues have found that attending to events in peripheral vision substantially decreases your capacity to report changes in the center of vision. But what this shows about our experience of the center remains up in the air. Is there no experience in the center in such cases (Mack & Rock's opinion), or is there experience but non-veridical experience, or...?

Daryl: I don't understand your first comment. Could you elaborate a bit?

Daryl said...
This comment has been removed by the author.
Daryl said...

By way of clarification, I would ask you to consider the visual limit-performance for inscribing what you call the space of clarity. You define it as the ability to identify a playing card's suit somewhere within a visual arc, but it could just as easily be defined as the ability to identify just the shape of the object, or just that there is an object present at all. In the other direction it could be defined as the ability to identify the finer details beyond just the playing card's suit. So I hope you can see how the space of clarity could shrink to an indefinitely small point, and how no amount of combination of spaces of clarity could produce vision. Vision is not combinatorial in that sense. In fact, spaces of clarity is just a metaphysical substitution for sense-data. They're on par with clear and distinct ideas. At least that's what I think at the moment.

Eric Schwitzgebel said...

Right. Surely, there's a gradation from relatively clear to relatively unclear. So seeing something like a playing card's value is in some sense an arbitrary test. But I don't see how that seriously undermines the question, or brings in metaphysical difficulties. I could ask: How broad a space do you visually experience with the kind of clarity that normally would be sufficient make out the suit and value of card held at arm's length? (This slightly awkward way of phrasing the question is intended to reflect the fact that the quesiton is not about our actual visual capacity to detect outward things but rather about the phenomenology of visual clarity.)

Daryl said...

If the question is about the phenomenology of visual clarity, then I don't understand why you need to ask about visual space at all. You either make out the suit and value of the card or you don't. Phenomenologically speaking, this places the card itself front and center, not whatever it is your eyes happen to be lined up with. When you talk about visual space you're assuming a pure, self-identical presence (like sense-data) when in fact there are only infinitely differential points of interest.

Eric Schwitzgebel said...

Shoot, Daryl, I'm just not understanding your concern. Are you denying that there's a fact of the matter about whether in visual experience there's a center and a periphery, with a greater clarity toward the center? Or are you saying that that's *not* the phenomenology of vision? Or...?