Monday, September 08, 2008

What Is It Like Not to Notice a Typo?

Most of us (certainly I!) can a dozen times read something we've written without noticing a typo. What, I wonder, is the phenomenology of that?

Suppose I've written a sentence with "that" where "than" should be. Maybe I've been reading Nichols on disgust, and I write "Drinking five glasses of saliva is worse that drinking one." What is it like for me to see that sentence as I read it? Do I see it at all (Hurlburt thinks maybe not)? Supposing I do visually experience the sentence, do I see the "that" as a "than", so that my visual experience, in the appropriate place, is actually "n"-ish rather than "t"-ish? Or is my visual experience really "t"-ish in that spot, though I fail to notice the error? Or is my visual experience somehow indefinite between "n" and "t" (and maybe some other shapes), even though I may be foveating (looking directly) on the "t"?

Suppose I'm also saying the sentence to myself in inner speech as I read it. Parallel questions arise. Do I utter to myself "than", "that", or some more indefinite thing?

Now my own hunch is that I see the "t" (or maybe something more indefinite) but utter the "n". At least it's hard to imagine that I would utter the "that" aloud without noticing the typo. But this is only a hunch, and I'm not a great believer in introspective hunches. Maybe in some future neuroscience, if we can narrow down more precisely the correlations between brain states and conscious experience, we could scan the visual system for a "t"-ish or "n"-ish representation in the right part of the visual system -- but that's a long way off, if ever we'll get there. Could more careful introspection get us the right answer? That's tricky, too -- not noticing something is necessarily an elusive sort of experience. It's hard to believe, though, that something as mundane and nearby as this would be beyond our ken....

18 comments:

Karlo said...

As someone who studies cognitive science, I can tell you that empirical studies show that most of us don't even scan all the letters of the sentence even after multiple readings. People experience much less of the outside world than their brains tell them they do.

Anonymous said...

I wonder if you have received the email that has gone around for a while. The letters in each word in the email are scrambled, saving the first and last letter. The thing is that it is surprisingly easy to read the message.
I know that I do not see many of the words that I read. And when reading anything besides philosophy I don't read many of the word.

Eric Schwitzgebel said...

Thanks for the comments, folks.

Karlo, I know that skilled readers only saccade a few times across each line. Is that what you're talking about? If the saccades were far enough apart that it wouldn't be physiologically possible to see most of the letters on the line clearly, that might help settle the question. My impression is that that is not the case, but tell me more....

Anon: Yes, that's a pretty cool message. Here's a Guardian article about it.

holyoke said...

It seems - and I apologize if I'm simply repeating everyone else's point, or missing somthing more subtle - that this is more of a pseudoproblem than a problem: missed typos are hard to account for given a set of assumptions about how perception works, but the assumptions are false.

One of the assumptions seems to be some version of the view that we see each letter individually, and construct each word out of them.

It reminds me of another issue in perception... someone more clever than me can think of a more appropriate example... but it makes me think of the retinal image, and how it's upside down, in relation to the things it's an image of. And an elementary mistake in psychology is looking for the part of the brain where the image gets flipped "right side up." But there is no such place.

holyoke said...

extra points if you noticed my (entirely accidental) typo...

Eric Schwitzgebel said...

Thanks for the comment(s), Holyoke! But what is visual experience like? Do we see each letter individually? And what exactly would it mean to say we do or we don't? I don't think these issues are at all clear! The uncertainly around these issues is part of what I meant to be conveying in the post.

I hope being confused about this isn't as bad as being confused about where the retinal image gets flipped to create right-side-up experience. Dennett might say these questions are confused in that way, but his view seems to me to be buried in self-contradiction.

Anibal said...

Word-letter advantage phenomena (people are more ready to detect words than isoleted letters)and other interesting phenomena of the pattern recognition mechanism of our visual systems are full of mysteries to investigate human phenomenology.

What if our mind/brains are constantly "guessing" about what is out there and we do not have a faithfull representation of the visual world, but rather an aproximate derivative?

This would be a mayor blow to "realism"

Perhaps the fact that we sometimes experience something instead of other ("that" as "than") is due to the probabilistc encounter with things.

We are more accustomed to see a fork in a "kitchen enviroment" that in a "forest enviroment", and here enters our cognitive machinery that abstracts the semantic gist (semantics about the wholeness not the details) of scenes, in this cae a text.

And if we a re reading a moral psychology paper that prime in us about what is like to be in "disgust" perhaps we are prone to read (to see) an hyperbolic semantic emphasis about "five glasses of saliva" instead of one.

Eric Schwitzgebel said...

I agree, Anibal, that it is likely that our visual experience is affected by our expectations. Kuhn and Hanson and Polanyi made this point nicely in philosophy of science around 1960, and the plausibility of this view is one of the reasons "sense data theory" and "the given" are out of favor these days.

I don't see, though, how it's a blow it is to any of the "realism"s in contemporary mainstream philosophy. What do you have in mind?

Anonymous said...

I don't think "infallible perception" is a requirement for realism, even naive realism.

Anonymous said...

Also, what is it like to make a simple addition mistake? E.g. when I was adding up (by hand) the different figures for my gross income on this past year's taxes, I got the same figure three times in a row. Then when I was finished with the form about 15 minutes later I went back and rechecked. I was off by a thousand bucks.

This doesn't strike me as a case of mistaken visual perception, at least not in the same way as with missing a typo. But something elseabout the two cases seems similar. As soon as you realize you mistake, you get a feeling of complete remoteness from your former self who was in error. (Or is this just me?) You can't for the life of you fathom what is was like to have been in that other state.

Anibal said...

About percpetual experience, that i think is deeply theory-ladden, i´m influenced by current developments in visual science (change blindness, inattentional blindness, filling-in phenomena)that people like A. Nöe borrow and bring to the philosophy of perception to examine the fiability of our visual consciousness.

And what is imply is that we have to be skeptic about the world becuse seem to be a grand ilussion, not only from the perspective of direct realism (if we represent faithfully what is out there) but also from subjective realism (we have to distrust our own "lookings" and "seemings" of things)

In other words, prior knowledge affect the way we see things and the mayor blow for realism will be if this prior knowledge (encoded in our genome?)is not coming from the "reality" that a visual creature must navigate.

Eric Schwitzgebel said...

Anon 7:37: That's an interesting comparison. Thanks for tossing that into the mix!

Anibal: If "realism" means a tight relationship between the world as experienced and the world as it is in itself, then I agree with you that this sort of case might be one (minor!) part of a case for antirealism, at least if interpreted in a certain way.

Badda Being said...

What is it like not to notice a typo? I would say that it's remarkably like reading something that's not a typo. Or like running a red light. Or like jumping off the Brooklyn Bridge.

Really, I have never understood the allure of these "what is it like" questions. Such questions beg for a comparison but provide no context for determining an appropriate answer.

Eric Schwitzgebel said...

Well, Badda, some people find the terminology helpful! I think you're right that something has gone wrong if you try to construe it comparatively. It's really just intended as -- at least I intend it only as -- another way of getting at "conscious experience" or "phenomenality" or "qualia" or "subjective experience".

jge said...

I think Badda is right: The question "what is it like not to notice a typo?" must be answered with: it's like there is no typo. This explains the alienation anon 7:37 tells us about. When we recognize the typo we wonder why we haven't seen it before. the latter experience changes the former, or maybe I should say: we feel that the latter should change the former but of course it can't, only our retelling of the experience

Eric Schwitzgebel said...

I think Badda is right, too, but I read him differently. I read him as suggesting that the act of comparison suggested by the "like" terminology is misleading.

In a sense, yes, it must be like there's no typo. At least in some respects. In every respect? Well, maybe, maybe not! And what it is like to read in general, typos or no? -- or to avoid the "like" phrase, what is the phenomenology of reading? The issue has hardly been explored, and different people give very different reports....

Badda Being said...

I wouldn't say misleading. I would rather say that the "like" terminology invites us to be led somewhere but without orienting us as to how we might know when we've actually arrived.

Re: your question "What is it like to read in general, typos or no?" -- we could say that it's like listening to an audiobook, or like walking, or like getting dressed in the morning. If you were tell me how any of these answers are incorrect, we'd actually be getting somewhere.

I think if you were to examine the notion of phenomenology you would see that the question "What is the phenomenology of reading?" really means the same thing as asking "What is reading?" But I wonder if you like to throw that extra word out there just as an obstacle to those who aren't familiar with it -- you know, in defense of the supposed difficulty of the problem you're trying to raise.

Eric Schwitzgebel said...

Thanks for clarifying, Badda! I sure hope I'm not just throwing up an obstacle with the word "phenomenology". I'm committed to the idea that philosophy can and should be clear and accessible. I dislike jargon, so I prefer to say "conscious experience" -- but since that phrase isn't always clear, it also helps sometimes to put the point in the jargon, too.

I think the question "what is the phenomenology of reading?" is a narrower question than "what is reading?" If I ask the latter, one part of the answer could be to appeal to all kinds of nonconscious processing mechanisms of word recognition, classification of letters, etc. The "phenomenology" question is meant to focus our attention just on the conscious part of reading, on questions like, "Do you visually experience the page before you when you read? If so, how much of the text is clear at any one time?" and "Do you have inner speech, or auditory imagery, of 'saying the words to yourself' as you read? And if so, is each word clearly enunciated in inner speech or do you slur and skip?" and other questions in that general vein.