view entire entry:
2002_02_22:23: Dream Sequences
to see pictures, view entire journal
(previous in thread, caro) :
~ indicate that we remember learned material much better if there is
~ a period of sleep intervening between the time of learning and the
~ time of testing. Why would this be so? If my emerging theory is
~ correct, it is because we need to shut down all systems not
~ directly essential to learn (or surviving), and focus all resources
~ on going over the database, for a long, uninterrupted amount of
~ In large part, the bizarreness of dreams provides the inspiration
~ for this emerging theory of intelligence. They give us some insight
~ into how we're doing this thing called knowing.
In email, Brandyn introduced the issue of Long Term Potentiation (loosely, "the permanent strengthening of synapses which occurs in response to bursts of input activity happening just prior to the neuron firing,") and suggested an alternate explanation of why we seem to need to sleep after learning something:
Because LTP takes 8 to 12 hours to have an effect. The learning is triggered at the time of the event, but the results cannot be used until the physical changes have had time to take place. This is established to effective certainty.This is useful information about the physical substrate. It's not fully integrated yet. Preliminary thoughts: While it is possible that those 8 to 12 hours needed for the physical changes to take place are usually happening while one is sleeping (the length of time is eerily coincidental, isn't it?), it is probably going on to some lesser extent while one is awake too. But it doesn't seem, offhand, that if I learn something in the morning then I'll be ready to use it by evening; more likely I'm ready the next day--a full 24 hours later, after I've had a night's sleep. So maybe the majority of this is happening during sleep.
But this is all very, very loose. Before getting too deeply into the discussion of how LTP is functioning, I'd want to establish what exactly is meant by 'learn', and by 'use'. For that I will need examples, of course. I need some article references.
Clearly, I do use some information as I'm learning it. A good example is a new step aerobics routine with new tricks with new names. There's only a minute to learn it and trying it; by the end of the class I'm supposed to have mastered it, and I could go home and teach it to a friend right away. It goes like this: At the beginning of the class, a new move is shown once, but we aren't shown the whole new choreography into which it fits--we're expected to listen closely and respond appropriately to increasingly complex sequences of commands. After we're shown the new move, we go on to the warmup and then to the beginning of the new routine. The new routine generally has moves in it we already know by name; they're just in a different order that we don't know ahead of time. Then, the leader reminds us of the move that we were shown at the beginning and expects us to do it in the middle of the rest of what we're doing. She deliberately makes up new moves that will stump us; she knows that's why we're coming back to her class, and that we'll complain if she doesn't do it.
Now, obviously, we're better at it the longer we do it. We are also better at it the next day. We get better at it if we visualize it between sessions. But nevertheless it does seem that we are using it right away.
Another example: An interactive programming class. The teacher shows us some new mind-boggling object. Then we do a couple of examples together, shouting out the next thing she needs to write on the board. Then we're turned loose at the terminals to write a little program to solve a problem using this new object. And we do it.
The contexts in which I've heard ordinary people talk about needing to sleep before taking a test are those where it is fairly clear that this material will NOT be memorizable in the hallway before the test. (Actually, some of us can pass a test with just-learned material, in a cheating sort of way, because we have a mental picture of it that we can then read during the test, but nevermind that for now.)
Here are some things that 'learn' could mean in the current context:
1. Commit to permanent memory at least to the extent that we can talk about it later
2. Commit to permanent memory to the extent that using it becomes second nature, in the way that, say, using the plus-sign when we are asked to make a program add is second nature.
3. Get the general idea of.
4. Have a conceptual overview of.
5. Integrate deeply into one's full context of knowledge.
1. Regurgitate for the test.
2. Apply to new contexts without cues such as a test that is obviously dedicated to testing it.
3. Talk about it interestingly.
There are surely other senses of these terms that it would be useful to distinguish, and possibly use distinct words for.
Second alternative to consider: See my essay on dreaming. [http://sifter.org/ brandyn/Dreams.html]I've had several days now to consider your explanation of dreams as risk-free practice. Brief overview: It doesn't seem prima facie unreasonable, but I do think it is too narrow (that just means that I think it fails to explain some typical and paradigmatic cases of reported dreams). I don't have an opinion yet as to whether it is true of at least some dreams; because it is so specific, I think it will take some time and effort form an opinion on that. However, I think it is compatible with the much more general theory I'm exploring. Let me re-read, and then discuss how....
...OK, I've re-read. In a moment I'll go out and deliver more brochures, and that will give me time to receive wide-ranging stimuli that will assist me in thinking about it. Here are my current thoughts.
First, I still don't want to talk about rat "dreams". I want to talk about what persons say when they wake up, and about what goes through my mind when I wake up.
Second, I need to try to think of actual dreams I've had that would fit the practice model. I am not sure this one fits, but it was a dream I had this week, after my conversation with Mike Young in which he offered the Pringles-can metaphor for a stack. Later that evening, I watched television show "The Guardian". In the story, a teenage girl was telling her child advocate that she and her younger sister didn't really have an address growing up, that they and their parents had lived in a mobile home, parked under a bridge. The next morning, I awoke and considered the following dream as I lay in bed: I "had been dreaming" that I was under a bridge, talking with some people in front of a table that had papers spread out on it. There was a mobile home in the background, also under the bridge. I had a bright red Pringles can in my hand, and I was explaining stacks to someone, I know not who or how many.
It is a challenge to see this dream as practice, unless by 'practice' we mean something much looser or broader than what I usually mean by it. Paradigm cases of practice, for me, are playing a harpsichord piece a couple of times trying to get the rhythm right; singing a difficult aria, in German, trying to do a better job each time to get the sequences of notes and foreign syllables right; repeating a new way of turning on rollerblades.
Less paradigmatic cases, which fall under a looser sense of 'practice' are, for me, deliberately getting into philosophical conversations with people who disagree with me so that I can find and be ready for new relevant objections. A lot less paradigmatic, to the point of overbroadening the concept PRACTICE, would be simply reading, watching television, talking, futzing with plants for no specific reason--in other words, practice as simply engaging in. Also, there is a sense of 'practice' that just means 'doing it or being that way'. It is in this sense that I say that engaging in _lying_ is a kind of practice, in that the more you do it, the more naturally it comes to you--more than mere habituation (which is just learning to tolerate, or something like that), but not exactly something that you're doing in order to get good at it, but you nevertheless get good at it by mere repetition in different cases.
Your case with rats can easily be interpreted as practice, for the obvious reason that you created it in order to show how dreams could be understood as practice. My case is less easily understood that way--not that it can't be, I'm just not seeing it. I feel like I have to leave a lot out of the dream on your explanation, that my explanation nicely encompasses. This doesn't tell me your view is wrong. It tells me it isn't enough. What is it that I am practicing, in this dream? Whether it's best to bring a Pringles can with me when explaining anything? When explaining stacks? That the can color matters (it is one of the most memorable and striking things about the dream)? Maybe it goes like this: Being under a bridge with a conference table full of papers in view of an RV and holding a food source is one scenario in which I might find myself explaining the philosophical background of the development of artificial intelligence. How did that go? Would it have been better to have been in the sun? Without the RV? Out of sight of distracting pieces of paper? In an actual conference room?
These are not rhetorical questions; I don't mean to imply by asking them that the answers are all "No!" It's possible that in fact the answers are all "Yes, these are factors which, during the rest of the night, I varied, and I was able to some degree to determine that explaining things to people inside of a classroom with a blackboard and desks for the students is in fact the best way to teach."
Another difficulty is what I am practicing in this dream. Explaining? Standing? Tolerating being in uncomfortable, visually unappealing places? Having the foresight to carry food sources around with me in case I find myself stuck under a bridge? All of these?
One answer to all of these difficulties is that I simply don't remember enough of the dream to make sense of it as practice. That answer is ok, I guess. But it does leave us with the difficulty of scant evidence. Some dreams are much more elaborate than this very sketchy one, but they don't necessarily lend themselves to interpretation via the practice model. For these, my model comes to the rescue. Would you be satisfied with that result? Or do you believe that the only explanation for why dreams exist is that they are simulations in which we practice, and thus every dream has a perhaps-inaccessible explanation in terms of practice?
A difficulty I would have with this requirement is that I tend to mix the most bizarre things in my dreams so that it seems extremely inefficient (in terms of speed, not in terms of aesthetics) compared to my own model of just going over the database to find likely fits.
Your model seems to focus on the question, "What if the world were like this?" In the case of the (fictional) rat's dream, the rat finds very relevant phantoms in his repeated resettings of the simulations: food smells, kitchens, cats, food rewards, hiding places that he knows. In my simulations, on the practice model, I encounter ever so many things that, given whatever you think is being practiced, are just irrelevant. Why would I need to practice those? Especially after almost 39 years of living on the Earth, I would think that I'd know better than to put unlikely things into my simulations. If it's a teaching simulation, say, you'd think I'd know better than to choose a dank shady underbridge with no place for the people to sit or even stand because there's a big RV in the way--maybe as a baby I'd consider that, but if all these years I've been practicing both while awake and in my sleep, wouldn't I by this time know better than to consider such silly things? Isn't that a huge waste of processing resources?
There's another dream I had recently. In reality, I had walked around the North side of Mount Soledad dropping brochures off at people's houses. I found one very large piece of ground with an empty house on it. I walked all around the house, feeling just a little out of place there but very curious to see if there was a chance that the owners were just away for the month and might be delighted to come back and find my brochure on the picnic table. The next morning, I awakened feeling like I'd dreamt about that house. There was a very deep, very steep "stairwell" that was most definitely not there on the real property. It was a sort of natural stair case made of roots and dirt. The stairs by the look of them seemed to be a navigable distance apart. As I bounded down them, the distances seemed to double, until I was flying down the staircase rather than bounding down it. Was I practicing going down stairs that change the distance between them as I traverse them? Or was I learning how to climb down a mountain that I had underestimated? If so why would I be doing that? Though I used to hike a fair amount, I haven't gone mountain climbing in probably two years; and while the terrain of La Jolla is hilly, I hardly need anything more than basic walking skills to navigate them, and I have those already. Was I preparing for the worst? For the big earthquake? I admit that this is a possibility. Maybe I'd got to the end of my stack, and now it was time to hang out admiring my database and making new connections, trying out new amusing things such as going to a stately home only to find that the "stairs" are really a huge dangerous vertical cave, and seeing what would happen if, in defiance of the laws of physics, I flew down it. I mean, why not? If I'm done with the rest of my processing, why not play?
My "grand" explanation at this point, which encompasses both our theories, is this: one of the many things that cause the subjective experience of dreams is the running of simulations to enable us to practice. But the subjective experience of dreams is also caused simply by waking up with a bunch of weird piece of data open because we happened to be in the middle of scanning the database. Those dreams can't be explained in terms of practice, and that's ok.
Perhaps, the more "sense" a dream makes, the more likely it is that we woke up in the middle of a practice simulation. But since the vast majority of reported dreams don't make as much sense as your rat-practice dream, I venture to suggest that practicing is only a small part of what we're doing while asleep. We're looking for connections in all sorts of ways, and one of them might be practice.
Another explanation that is less friendly to your practice theory might be this: We report dreams fairly selectively. Maybe we're just leaving out all the stuff that we can't tell a likely story about, and grasping at the few elements of which we can make sense, in just the way that commissurotomy victims do, and making up a likely story just about those. I think this is what my own experience of dreaming is like; I tell a story that's the best I can do, and just leave the stuff out that's too weird. This is illustrated also in the answer to your next question:
> I have a question-by-example for you that would help
> me better understand your theory here. Take, for instance,
> Can you elaborate what you think the actual process was
> which created this memory upon waking?
I think so. Let's say at the point at which you got to the Church of Satan in the stack (which I seem to recall we were discussing right around that time, November 2000?), you were comparing it to some acrobatic images or moves or some figure skating stuff which were tucked away in your database. With evil sadistic charismatic people you also have some associated data such as religious showmanship, possibly the Steve Martin movie about same, supplying you with things like the wireless mic and the humor. It's hard to speculate without asking you more questions about what you know and what you've seen.
Let me try answering your question another way. Let's say that this had been my dream, not yours. Now let me explain it using my theory. I had been reading an essay on objectivism on the church of Satan site, before the dream happened. I'm scanning the rest of the database, and I come across the Cirque du Soleil data. Those two things seem similar, in that both are, to me, really dark (literally and figuratively), and weird, and mysterious in presentation, and intended to have a mystical quality about them. A kind of match. So I open up more of the details of the data about Cirque du Soleil: acrobatic, death-defying, rather bizarre stuff even when you're not dreaming. These performances happen on stage. What are some other things that happen on stage? The soup points out that church events happen on a kind of stage, and so do big university classes; in fact, when I teach, I feel like I'm giving a performance, and expect pretty much anything to "happen" (statements and questions from students are highly unpredictable and often bizarre): more similarlities. And universities figure big in my life, because I've done lots of good stuff and met lots of interesting people there. There are real possibilities for integration here! Let's load this whole chunk of circus and performance data into the temporary hash for further use somewhere else in the upcoming parts of the database.
Now I awaken. I've got three basic things in my hands: weird, dark, elaborate visual images of Cirque-du-Soleil-type acrobatics, (2) the Church of Satan (from the stack), and (3) stage performances broadly-construed. I awaken with a weird sense of all these elements being there togehter. And time being nothing but the distance between the ideas passing through the mind, as John Locke would have it, and the mere processing of the data probably having at least some sense of time associated with it, I immediately apply the dimension of time, lay the "events" out flat to the best of my ability on a time line. And that laying out "in time", maybe, is nothing more than listing the bits of data that I happened to have on my clipboard at the time, and lists having a temporal interpretation, I get a little story line that manages to make a bizarre kind of sense of data that really doesn't make any sense. Dreamatic license, we tend to give ourselves, and each other.
Now you try!
And I'll try your theory of practice: This dream is in some ways easier to see as practice than my Pringles can under the bridge dream. I am a physically active person. I also like to watch amazing feats of arial acrobatics. I'd like to investigate what it would take to do some of that stuff, safely. So I run a simulation that I can watch.
That's where the application of the practice model stops being easy. It's the Church of Satan and its minions that gets in my way here. Perhaps I wake up in the middle of simulation Number 50, which happens to check to see whether devil-worshipping would be helpful in developing acrobatic skill; I'd already run simulations that checked whether it would be better to be a rat, or a dove, or an objectivist, or a priest, or a space alien, or a software developer. Or maybe I just like to let devil-worshippers do all my stuntwork for me, as a sort of cosmic justice; and when I finally get a picture of the moves in my mind and I know it is safe, then I'll jump in and do it myself? Maybe the reason that I'm not doing the moves myself, is because learning by watching is a skill I need to practice too?
I feel like there are important aspects of the report that I have to leave out, in order to make it a practice simulation. Or else I leave them in, and find that I am, well, wasting valuable processing power and time by having weird irrelevant characters doing my simulations while shouting out quite strange and philosophical cues. Or maybe that's just for humor's sake? (Humor figures big in my intelligence theory, so I wouldn't be surprised by its coming in here.)
view entire entry:
2010_04_28:11: looping and intelligence
to see pictures, view entire journal
[posted to a discussion list this morning]
A couple of days ago I suggested a strategy for coming up with a definition of intelligence. I started with a list of beings that I consider intelligent and asked you all to contribute to it. So far on our list are
Next step would be to ask what it is that all those entities have in common. Instead, some of y'all are continuing to focus on who is smarter than who, who is not as smart as YOU, and what sorts of behaviors you find dumb. That's a fine project, but it's tangential to mine. You've gotten yourselves into an infinite loop; given that this is a philosophical discussion where philosophical rules of procedure are not being followed, it's no surprise, happens all the time. That's ok, if you need to get this out of your systems. I will go along with your tangent for a (rather long) moment. Hopefully I can demonstrate why it's a better plan to go back to the list of so-called intelligent entities and ask what they have in common. Don't let me stop you from engaging in your project, but I invite other people to continue with my project simultaneously.
Some examples came to mind while reading about how to avoid looping behaviors. At the end of my list of examples I discuss why I don't see these as problems but rather as more data about intelligent beings and how to create them.
1. Person loses something. Person knows item was used last night, in the house, after coming in for the last time; so it must be in the house. Person searches house for item. When all possible locations for thing have been checked, they are all searched again. And again. And again. With increasing feelings of frustration. Feelings of stubborn determination increase with the feeling that, since so much searching has already occurred, person must now be very close to goal. Other members of the household are interrogated and enlisted in the search; members question person about all the places person just searched, just to make sure. This can go on for hours, or days, or WEEKS, even when it might be more economical to just buy a replacement item (like a pen, or a sock).
2. Person wishes to purchase a particular piece of clothing: a leather jacket, say. Person has standards about style and fit that are at odds with current fashion. Several pretty good jackets are observed, but none are the Ultimate Perfect Jacket, and so are passed up with a mental note that if no better is found, person will return to buy this one, EVEN IN CASES where the jacket is on closeout and is likely to be gone before return. Every area mall is searched, store by store, even where clothing is "organized" by label rather than by function (e.g., Nieman Marcus and Nordstrom, where they might stock 50 different leather jackets, but they are distributed over approximately 10,000 square feet and hidden amongst the ties, underwear, shorts, jeans, slacks, hats, suits, etc. Stopping for lunch or dinner would be unthinkable: hunger is irrelevant when the Perfect Jacket could be just around the next corner, purchase to be followed by celebratory dinner. Search on foot only ceases when malls close. Next trip includes long drives around city looking for leather stores, motorcycle stores, fashion boutiques, etc. Upon encountering leather store, every likely jacket is tried on, twice. During hours when store shopping is impossible, the internet is scoured. This search continues for two months. Just a little further, just one more jacket, just one more mall in one more suburb...
3. Boy invites girl on "camping trip", a three-week touring vacation to be taken on the way to a conference. Anticipating, girl chats happily with boy about cooking over and warming selves next to open fire, listening to night creatures and chatting about philosophy, then waking up and taking long wilderness hikes and swimming in mountain lakes, nerdily identifying birds with bird book. Boy and girl both pack novels, expecting to read lazily in scenic spots. Boy and girl jointly decide that route will proceed from Indiana south to Virginia, wind north along coast until destination in Toronto is reached. The first day, girl is not allowed to eat or go to the bathroom for 12 hours, because boy insists on driving as far as possible before camping for the night; tent is pitched in the dark, nuts and crackers are eaten for dinner. Next morning boy leaps out of sleeping bag at dawn and begins tearing down camp: 15 hours more of highway driving, one rest-stop is grudgingly visited. Girl complains of hunger, explains female need for frequent bladder relief, and reminds boy of desires to come in contact with wilderness. Boy listens, nods, then gets on highway and drives 15 hours, to get as far as possible before eating nuts and camping for the night. Following this pattern, conference destination is within reach by the third day of the three-week tour. Now touring can begin, says boy. Begin daily 12-hour drive back and forth across northeastern states, to get as much driving in as possible before camping for the night--every day, despite louder and more frantic complaints from girl, for the remainder of the three-week "vacation". Loop ends only when conference begin-date arrives.
4. Person is raised by alcoholic parent, observes irrational, risky, abusive behavior, swears off alcohol entirely, and vows never to consort with alcoholics in the future. Meanwhile person gets very good at putting up with, excusing, disguising, sympathizing with, and caring for alcoholics, a fact readily observed by alcoholics. Surprise! First lover is alcoholic. Person endures behavior similar to parent's, eventually ends relationship, puzzled by failure to recognize and predict problems with lover, renews vows to never drink and never go anywhere near alcoholics in future. Friends note that second lover is "just like" first lover, person resents friends interference, second lover indeed turns out to be alcoholic, person breaks up with lover and apologizes to friends... Repeat. Repeat. For years, perhaps for decades. See extensive literature on this phenomenon (or look around at friends :).
5. Person is rude, mean, and unfriendly. Person can't get laid. Person imagines that other people are deterred by person's frightening intelligence. Good friends try to help, tell person that it is not intelligence, but overall ickiness, that repel opposite sex. Person briefly considers shaping up in order to get laid, decides it would be "dishonest": This is who I am, I'll find someone who LIKES rude, mean, unfriendly people!" Remains bitterly single for long stretches. Eventually finds lover similar to self, ends relationship because person dislikes rudeness, meanness, and unfriendliness in other people. Tries being (dishonestly!) nice, gets laid, eventually drives lover away by reverting to repulsive behavior. Loop continues for entire lifetime.
I have never met a single human being who is free of this kind of looping in *some* area of his or her life. I ask myself what that means, what it indicates about the way the human mind works. Y'all seem to be looking at "lower" animals for examples of irrationally repeated behaviors, but if you try to be a little more objective, you'll see it all around you; if you're a lot more objective, you will see it in yourself, too.
I have no problem with the project of making a being that is "better" than existing human beings. But my guess is that, before we can do that, we need to at least be able to talk about how we would make one that is *as good as* human beings--or even as good as a sphex wasp.
When I see behaviors like the examples I listed above, or like the sphex wasp being tortured by irrationally repetitive entymologists, or tigers pacing back and forth in tiny cages, my first thought is, What is the underlying explanation for this looping? Why do I see it in every instance of every category of creature that I think is intelligent? My hypothesis is that the strategy is sound. It works well the vast majority of the time, however the given creature defines 'works well' for itself. And then in a few weird cases, it doesn't work well at all by anyone's definition of 'works well'.
Rather than say, "Look, this strategy sometimes fails! Creatures who fail this way aren't intelligent--or at least aren't as brilliant as ME! Let's make a creature that will never fail in this way!", I say, "Look, there is a reason that every intelligent creature, INCLUDING ME, sometimes fails in this way, yet succeeds in its projects most of the time--what is the strategy that usually succeeds but sometimes leads to this failure, and how can I distill the strategy down to its essentials and reproduce it, accepting that it will sometimes fail?"
(I anticipate objections: No, I'm not deliberately trying to build something THAT FAILS. I am NOT following a research strategy based on the principle "to err is human". After all, there are an infinite number of ways to get things wrong--how do I know that any random way of getting things wrong is just the way that intelligent creatures tend to get it wrong? Instead, I'm asking why something that usually succeeds sometimes fails in very specific ways. If I can build something that usually succeeds by a given strategy but sometimes fails in just these sorts of ways, then I'm doing it *right*. I suspect that a consequence of proceeding in this way is that I need not worry ahead of time about what I will do when I get into an infinite behavior loop. I've found that what usually happens is that, when I do it right, these sorts of terrible problems you are trying to hedge against either don't come up at all or are trivially handled by the agent itself.)
view entire entry:
2002_02_21:15: Dream Beginnings
to see pictures, view entire journal
responses (thread index)
next in thread: michael2002_04_02:21:24 (thread index)
next in thread: caro2002_02_22:23:17
New Topic for the journal. Here I'll write notes for my paper
on a.i., which I will present at the Second Annual Enlightenment Meeting
First, some distinctions need to be made to keep things clear.There are several terms that tend to get confused in discussions of dreams: 'REM sleep', 'brain wave pattern', and 'dream'.
REM sleep, or Rapid Eye Movement Sleep, is detected by looking at the sleeping person's eyeballs rolling around under the eyelids. For the purposes of this discussion, this is all that will be meant by REM sleep--that the eyes are visibly rolling around.
NREM sleep, or Non-Rapid Eye Movement Sleep, is detected by the absence of rolling of eyeballs under closed eyelids. (In the OCM article on dreams, Iswald says, "There are two kinds of sleep..." Surely much finer distinctions are in order?)
Dreams are detected by asking conscious, non-sleeping people who have just awakened (or by their volunteering), generally in verbal form, "what were you dreaming?" or "what thoughts were passing through your mind?" The people then report, verbally, a dream. I sometimes refer to this as the subjective part of the experience, the qualia, if you will, of the neuronal activity.
IMPORTANT: People will report dreams when awakened during BOTH kinds of sleep. The difference is that, when awakened during REM sleep, people report more visually-descriptive dreams; when awakened during NREM sleep, they report less visually-descriptive dreams.
My Old Theory of Dreams (which I still believe, and now may have a good foundation for, in my New Theory of Intelligence)We don't dream while we are asleep. Upon waking up, we confabulate: i.e., we tell a likely story that makes sense of the neuronal activity that had just been occurring. The origin of this theory is some information about pilots. When pilots stay up in the air for a long time, sleep-deprived and subjected to reduced air pressure, they report that they "saw" gremlins on the instrument panels. Sometimes the gremlins talk to them. Because the reports were so similar, coming from different pilots, the phenomenon was investigated. It was found that the capillaries in their eyeballs burst, causing some optic nerve stimulation but not in a way that the pilots could say they consciously saw. The conclusion was that, in order to make sense of this stimulus, they confabulated a story. Talked to Tom about this last week, who suggested that the reason that they pilots specifically reported seeing gremlins is that the term 'gremlin' has a long history in aviation. Anything that goes wrong in a plane is caused by "gremlins." Thus it would be a word in their ready-to-hand vocabulary, and some disturbance in their visual field could be interpreted as something "going wrong"--and since they were overtired (and possibly partly asleep), it wasn't all that surprising to them to actually start seeing the gremlins that were the source of all the problems. And this inspired me to suggest that, the human brain being set up to distinguish human faces better than pretty much anything, if the pilots had some visual sensation they couldn't immediately resolve in some other way, they might try to find a face in the blur. Gremlins. Completely confabulated in a more or less story-like way, on the basis of neuronal firing that couldn't be resolved.
This piece of information suggested that other bizarre phenomena might be similarly caused and explained. Thus, my theory of dreams, which recurred to me recently while I was drawing some diagrams in an attempt to understand and solve what Ted called "the binding problem in artificial intelligence". I'll detour here into my results, which didn't solve the binding problem (although it made me increasingly suspicious that it was a spurious problem akin to Xeno's Paradoxes), but produced a pretty little theory of how to save on processing and storage overhead when developing an artificial intelligence.
According to my developing theory of intelligence, we sleep in order to sort through and reorganize what we know. I refer, loosely and metaphorically, to what we know as "the big database."
Short-term memory is first in, last out: a stack. Like a can of Pringles Potato Chips, as Mike Young so handily put it when I was explaining stacks to him this weekend. Onto the stack, all day long, data gets pushed in the order in which it occurs: images, nice turns of phrase, people's names, faces--in fact, EVERY thing that we experience, whether we notice it consciously or not. Evidence that short-term memory can be usefully described this way, is this sort of example, which happens to people all the time: Suppose two people are having a long conversation. At some point they stop and say, "Wait a minute! What got us onto this? Weren't we talking about something else? Seem to remember it was important..." In general (and there are exceptions which I will consider separately, because I can use them to help explain the mid-term memory data structure)--in general, people find out how they got here in the conversation by literally retracing their steps, backward, thus:
"We started talking about libertarian politics because before that we were talking about Danielle Van Dam and how all of this could have been greatly alleviated if children had chips and transmitters implanted in their bodies and you started getting defensive about surveillance; and we got onto Danielle because I was half-watching the news while we talked, and, uh, before I got distracted by the news we were talking about, uh, Oh! Your ski trip, and the reason we were talking about that was that we were talking about the upcoming meeting and you were wondering whether you had the money to do both...This kind of retracing seems to literally go in order: last in, first out. I'm popping the stack, trying to get to the bottom of the Pringles can so I can see where this conversation began. It is difficult--though not always impossible--to do it in another order.
The short-term memory stacks and queues are where we put stuff for temporary storage. These structures are not for understanding, but for brute storage. Later, we will empty them out element by element, and check it against the database for differences and similarities.
The big database is a multi-dimensional hash, or, as Ted O'Connor noted this week, it is not really a hash, but more of a soup. My understanding of a soup is that, like a hash, there are keys with associated data, but the structure is multi-directional. This avoids the wasteful problem of having a key refer to data that refers to the same key further down in the multi-dimensional hash. Instead, the data refers back to the key. I don't care how precise this is right now. I'm in a hurry, and there is much to be blurted out. My stacks are really, really full, my processors are working overtime, and I may even be asleep right now. I think I hear them coming to take me away....
Mid-term memory, which may come in several degrees or sizes or structures, is basically a temporary hash separate from the big database (where 'separate' might even mean, 'is one special flagged element of the big hash'). Recently, Ted and Jing Jing were here. Ted started some subject, Jing Jing asked a question about it. Though the subject was mostly unfamiliar to her, she suddenly said, "Oh! This is what Bernard was talking about." and then had something to say about what he said. Where was this stuff? It wasn't in short term memory; it didn't seem she was digging it out by popping a stack. Rather, she listen long enough until Ted said a key word or phrase--and it was a key she had in her mid-term memory hash. The data associated with that key was, in part, Bernard, and what he'd had to say about the topic.
Mid-term memory is where we are putting information that we need to access in a flash. Sometimes this is stuff we have already found places for in the big database, but we've pulled back out
Mid-term memory is where ALSO we are putting things, sometimes while learning them, that seem important but we don't yet know how to fit them into the big database--or perhaps even where we put things when we know how important they are but we're too tired to make a big, expensive query to the database (and we might say "I don't feel like talking about this (normally favorite) subject right now"). We forego brute efficiency of the stack and take on the higher space overhead of the hash in order to benefit from the random-accessibility of the hash. We can extract the information with lightening speed, even when we don't yet have a full understanding of it in the context of our wider knowledge. Perhaps the rule is even, "If it happened recently, it must be important." or "If it happened recently, then there's more information related to it that is sure to follow." Incidentally, this is another way of putting Price's classic theory of concepts as dispositions, where some concepts are just on the threshhold and ready to be used whereas others are buried more deeply and need more stimulation to be activated.
Article Analysis: "Dreaming," by Ian Iswald (professor of psychiatry), OCM, pp 201-203
Let's look at some of the prejudices that are apparent in the article, simply stated without argument as basic fact.
The first sentence: "In our sleep we all intermittently experience insanity" (Dreaming, p. 201). This provocative sentence indicates that the author thinks there is no reasonable explanation of the bizarre things that we dream. I think that there is.
A couple of paragraphs later, Iswald says "We may need to dream, but no one has yet devised an experiment to see whether we have such psychological needs at night" (Dreaming, p. 202, col. 2). This statement indicates that, because the reported dream is considered a psychological phenomenon, the author's own thought processes are focused on the dream as psychological and important in and of itself, rather than as an effect of something that is important. I think it is an effect of something much more important.
Another strange statement regarding data: "While the majority of the REMs certainly cannot be ascribed to scanning the visual field of a dream-world, there are occasional large eye movements that do seem to bear a relation to described dream content" (Dreaming, p. 202, col .2). This statement suggest that Iswald is firmly in the camp that the actual _dream_ occurs during sleep, that the dream itself is the significant, interesting activity, and that the point of the neural activity and rapid eye movement is to cause the dream, rather than to do something important to us.
I think that dreams occur because we wake up and, being explainers and shelvers, need to make sense of the activity that had been occurring in our brains.
More on this article later.
Sleeping, Neuronal Activity, and DreamsPopping the short-term memory stack and checking each element against the whole database is expensive, both in terms of processor overhead and time. Whatever else sleep may be for (healing, rebuilding, growing), this is when we are reexamining our information, finding fits, making connections, and storing. Studies indicate that we remember learned material much better if there is a period of sleep intervening between the time of learning and the time of testing. Why would this be so? If my emerging theory is correct, it is because we need to shut down all systems not directly essential to learn (or surviving), and focus all resources on going over the database, for a long, uninterrupted amount of time.
In large part, the bizarreness of dreams provides the inspiration for this emerging theory of intelligence. They give us some insight into how we're doing this thing called knowing. Consider the things you might store in any given day. Consider taking a single thing off the stack of things you stored, and checking it against the whole database. Suppose that I store the fact that Terry wants to go to the nursery to pick out the plants listed on my plan, on Saturday. I see him as he's telling me this, leaning over my patio gate. While sleeping, I get to this piece of data in my short-term memory stack, and check against everything I know. I'm somewhere in the medieval history portion of my database, when the alarm clock rings. I wake up with some disparate piece of data flying around in my head. I try to tell a story to make sense of it. So I say "I was just dreaming that Terry came over to tell me when we were going to the nursery--only I wasn't at my apartment. I was at the philosophy department at Indiana University. I was in Paul Spade's medieval logic class, and suddenly Terry just leaned in the window and started scheduling with me, and for some reason I didn't think this was strange at all...."
view entire entry:
to see pictures, view entire journal
Some Background on Laughter and HumorThere is no entry for Laughter in The Oxford Companion to the Mind, although there is one for Laughing Gas. The only entry in the index refers the reader to the article on humor. Initially I found this very strange.
But having read the article under the Humor entry, I think I now understand why there is no entry or index cross-referencing on laughter. It's because people, most especially the psychological community, are extremely confused about it, and would rather not deal with it, if indeed it ever comes up for them as an issue at all. In fact, judging by the very strange things that the author of the Humor article says and the even stranger things that the researchers he quotes are saying, it would surprise me greatly if any of them ever laugh. Let me draw a distinction, using my own terminology.
There is a difference between genuine laughter, and contrived laughter. Genuine laughter may be recognized in one of two main ways: (1) the character of the response, and (2) the circumstances under which the response occurs.
How to tell by character? You are witnessing genuine laughter if it is largely uncontrollable. The person who is experiencing genuine laughter seems to be out of control to a certain extent, unable to stop laughing voluntarily. If you never or rarely laugh yourself, or think there is something forbidden or problematic about it, it may be just as easy or easier for you to recognize genuine laughter in other people, rather than in yourself. But most people can at least dimly remember a rare time when they couldn't control their own laughter. If you are having trouble understanding what I mean, try watching "Whose Line Is It Anyway"? Drew Carey, the host, almost invariably collapses on his desk at some point in the show, tears streaming from his eyes. Listen to the sound of his laughter. That's what genuine laughter sounds like. Or, if you are old enough to remember "The Carol Burnett Show," you will have seen Harvey Corman and Tim O'Connor attempting to control their genuine laughter in the middle of a rehearsed skit; they frequently have to turn away from the audience to hide laughter that is leaking out anyway. The secondary symptoms of laughter--redness, tears, falling on the floor--vary from person to person. Sometimes, people run out of breath and can no longer make any sounds; they may lose all voluntary motor control and speech capacity, and may be capable of only bodily convulsions or shaking. The convulsions or shaking are the genuine laughter; sometimes this results in sound, but in severe cases sound is not possible. If you get close to the person, you might hear just a slight, jerky hissing.
If you really don't know what I'm talking about (especially if you have rarely laughed yourself and live amongst humorless people), then for you the circumstances under which the response sometimes occurs may be the easiest way to understand the distinction between genuine laughter from contrived laughter. Sometimes, the only time a person experiences genuine laughter is when he is being tickled. Even the most repressed, humorless person will engage in genuine laughter under these circumstances.
In the rest of this entry, when I refer to genuine laughter, this is what I mean.
Contrived laughter has various reasons for existing. The OCM article on humor lists many of those reasons; as far as I can tell, if focuses on instances of contrived laughter, not genuine laughter. The problem is that, in trying to explain humor or laughter, if you take all of these instances of contrived laughter into account, you get a big bloody mess that can't be sorted out. It is analogous to including under the concept PHILOSOPHY, every thought, conversation, comic strip, or dog's bark ever to come into being--no wonder that you can't make any headway. Contrived laughter is the kind that exists for the puposes of
...preserving order, changing group desteem and cohesion, expressing allegiances and revealing attitudes with relative impunity, testing the standing of a relationship, maintaining or undermining a status hierarchy...used to precipitate an absorbing and pacifying digression...to defuse the threats of others by engendering a debilitating discomfort...make light, or pretend to make light, of their own misfortunes and predicaments (Humor, 320, col. 1).There are other reasons that it exists too, but you can see the very negative focus of the article's authors and their references by scanning this list.
For the most part, the OCM article on humor seems to refer to contrived laughter, but it is hard to tell in any given paragraph. The authors do seem to have some sense that there is a difference between genuine and contrived laughter, but it is only implicit: for example, they say "It seems that humour loses much of its splendour, infectiousness, and power under laboratory scrutiny, to such an extent that exuberant laughter is rarely elicited from experimental subjects." Here, when they say 'exuberant laughter', I infer that they are referring to what I am calling genuine laughter.
Contrived laughter is, simply, everthing else that is not the genuine laughter I explain above. It covers a broad range of noises and gestures. My friend Heather taught her baby, Emily, how to engage in polite contrived laughter. Heather says Emily's name, and when Emily looks at her, Heather smiles and says (SAYS, not laughs), "Hah, hah, hah, hah, hah." Emily, who wasn't saying more than 2 or 3 words at the time I first saw this being done, then imitates her without smiling. Emily has gotten so good at this that she will offer this sound as a response when she doesn't know what else the big humans around her want her to do or say, or when she is trying to communicate something and no one understands her. Just like adults! This 'hah, hah' utterance is dramatically distinguishable from what her father calls "the war whoop," which is emitted loudly and does not seem to be a form of communication at all. Rather, like my own genuine laughter, it seems to take Emily away somewhere private, or at least to stand in between her and the people around her, signifying to them only that she thinks something is funny.
I once knew a man who punctuated every sentence with the sound "ughCH!" When I first met him, I thought he was making the sound because he thought (or meant others to think) that what he was saying was funny. Indeed, when he was clearly making a joke, he would bark, "ughCH! ughCH!" But it turned out that it was really just a sign of a full stop, the period at the end of his sentence. When I say that genuine laughter has an uncontrollable character, I don't mean that it is the result of entrenched habit, as is the replacement of full stop with a laughlike coughing sound. That may be to some extent involuntary after years of repetition, but the loss of control of which I speak is more holistic, full-bodied, visceral, and consciousness-gripping: the thing that is being laughed at occupies the mind and body to such an extent that nothing else can thoroughly penetrate or be attended to (in some cases, as is usually the case with me, the laughter and its uncontrollability become part of the thing that is being laughed at. My laughter is recursive, massively parallel, non-linear, and can frighten repressed adults. I consider this one of my better traits.).
There is also a strong tendancy in the Humor article's authors, and in many people I talk to, to associate a necessary social component with laughter and humor. Somehow, the laughter is for the benefit of others, or for the benefit of the group as a whole. They are referring in the main, almost certainly, to what I am calling contrived laughter. Genuine laughter does not consult the social melieu before possessing someone; it has a tendency to burst forth even at the least socially appropriate time. While it is true that, when I laugh in a group, the reactions of other people who are watching me will add to the humor for me, nevertheless I'm quite capable of the same laughter when I am alone.
Case in point: It's Thursday night. I am completely alone; my next door neighbor isn't even home, and my doors and windows are closed. I've just eaten dinner and I'm finishing up the second piece of pie and cup of decaf coffee. My stomach is exceedingly full--much more full than is comfortable or usual, but I couldn't resist the hot, freshly-baked strawberry pie. It is exactly the wrong bodily state to be in for uncontrollable laughter. In defiance of the great personal risk, I have the television on, tuned to "Whose Line Is It Anyway?," which as usual is making me smile. But one of tonight's improvisations is horribly, horribly wrong in every possible way: random audience members have been wrenched from their seats, handed microphones, and are now supplying completely inappropriate sound-effects for Collin and Ryan, who are at the amusement park despite the fact that they are supposed to be delivering their baby any moment; they don't bother making up a skit themselves, because the sound effects are so terribly wrong that it's funnier to react to those vaguely along the theme of inconvenient baby-delivery. I am laughing so hard that I am simultaneously regretting having eaten so much pie (and this predicament becomes funny too, of course). To prevent myself from exploding, I am forced to slide out of my chair and into a kneeling position on the floor. Thank god for commercial breaks, or I might not have survived it. I behave the same way among other people, if there is something funny enough, and I assure you it has little to do with their approval, my social ranking among them, my need to appease them, or my desire to foment group cohesion--on the contrary, I frequently run the risk of confusing and disrupting the group when I do this. It's not a desirable outcome, but that is the nature of genuine laughter: genuine laughter is for the laugher, not for other people. Are you with me now?
The Purpose of LaughterAs I said, the reasons for contrived laughter are many, complex, and partly but clearly not exclusively social. I'm not interested in those right now. I'm interested in the basic, mostly involuntary, physical response of genuine laughter that comes prepackaged with every normal human being. What could it be for? Is it for anything? Is its pupose merely a precursor to, and a primer for, contrived laughter--or, intersectingly and more broadly, for social relations?
Here's what I think. Genuine laughter is something that we need. Just like vitamins, or sexual satisfaction, it is something we may deprived ourselves of and still survive. But to really thrive, we need vitamins, sex, and laughter. This is my overarching opinion, but I'm not going to attempt to try to prove this here.
Why We Laugh
To talk about why we need it, I will have to provide a description of the psychological circumstances under which genuine laughter occurs. Since I have not done a study, and I know of no data collected (most people can't even make the distinction), I will have to speak about my own case and of the cases of genuine laughter that I have witnessed in other people. The discussion will therefore, of necessity, have a Caro-slant. If you haven't experienced or witnessed (consciously) instances of genuine laughter, you will be completely lost by this discussion.
I have thought for a long time about what it is, generally speaking, that makes me laugh. Several people's observations about me have helped me to clarify it. My own observation is that I like absurdity, and the more absurd, the harder I laugh. Mary Jane Padrone observed, many years ago, "You like humor that makes you imagine something. You like to picture it in your mind." I think this is true of me. It could be because I am a highly visual person (by which I mean I delight in the stimulation of my eyes, and if I am deprived of varied visual stimulation I am unhappy and bored and can't concentrate; I don't mean that is the way I "learn best"--I don't think it is). David observed that my laughter was "non-linear," by which he meant that I start off laughing in response to the ostensible humorous thing, but then my laughter changes in character and intensifies as I make more and more connections to things not visible to the observer as well as to other things in the immediate scene that suddenly become funny in virtue of the original humorous thing (this is very abstract, but I'm trying to get it all down quickly; I'll have to provide examples later). My old friend Tom Marshalek once said that my laughter frightened him, that I seemed insane, and that he almost felt afraid that I would draw him into my insanity. My sister, whose laughter is similar to mine, said that I was like a woundup spring.
David and Mary Jane may seem to have said something different, but I think the essential element here is that I am imagining or picturing absurdities, and that I am making new absurd connections as I laugh. Tom Marshalek thought he was seeing insanity because (as I remember) he almost never got the joke; I'm sure I never saw him genuinely laugh, though he engaged in loud contrived laughter.
In the past month, thinking hard about a.i., I have come back again to the subject of laughter. I think it all started when Tom Radcliffe asked if our a.i. will laugh. And this raised the question for me of what purpose laughter might serve in human beings.
I have long believed that laughter is extremely beneficial. I know that my life is better, the more often and the harder I laugh. How I loved working at Designscape in Bloomington, Indiana! It was miserably hot under the summer sun, horribly humid; my normal body temperature being only 96.8, things were even worse for me than for ordinary human beings. The work was physically difficult and tiring; when we had to prune by hand an entire hedge of yews, it could be deadly dull. We were paid just a little over minimum wage. But I loved it, because I laughed all day. Mary Jane was by far the funniest person I have ever known. In a flash she could sum up a set of circumstances or make connections between disparate pieces of information and put her own Mary Jane spin on the result; and in a flash I would be rolling on the ground trying to breathe. Truly, working outdoors is healthiest for me, even in the unendurable heat of Indiana; but laughing all day long made me feel physically well and very happy. If it weren't for all the laughter, I wouldn't have continued to work under those extremely harsh and unpleasant conditions. So not only could I put up with difficult circumstances, but I felt really good too.
Aside from making me feel good, what else might it be doing for me? Why was it that the absurdities were what caught my attention and made me laugh? What do other people laugh at?
I think that the common element in the things people think are funny is wrongness. There is something wrong with what is being presented. It is most emphatically not the case that all wrong things make people laugh, or that the same wrong thing will make everyone laugh. (Your reading comprehension will be tested on this point later.) What one considers wrong is completely dependent upon one's own personal knowledge and current circumstances. This seemed a very broad conclusion to make about humor and laughter, a month ago, so the first thing I did was go over to Heather's house for more information. I asked her what makes Emily, 18 months old, genuinely laugh. "I mean, really laugh, not that little fake social laugh you taught her." Heather thought for a moment. "She really laughs when you run." Now, that surprised me! Of all the things that a baby could laugh at, why would she laugh the hardest when someone does something so ordinary? Heather could see the disbelief on my face. So she showed me. Without ceremony or comedy or comment of any kind, she simply called Emily's name and then ran once acroos the livingroom. Emily did indeed laugh, her face reddened, her eyes squinted, her mouth opened in a huge happy smile, and the sound came from deep within her. About 10 seconds after Heather stopped running, Emily stopped laughing, but she continued to smile happily, ehthralled. Heather then ran back a forth a couple of times. Emily was in hysterics. Remember, this is the little girl who already knows how to contrive a laugh; and that was not what she was doing. She was not offering a comment or a communique; this laugh was for her and her alone.
What good could this possibly do her? If I am basically like other human beings, and my hypothesis that it is things that are somehow wrong that make me laugh is applicable to other people, it should be applicable to Emily. But it's hard to see what is wrong with running, a perfectly normal animal thing to do. But is it normal? For Emily, it isn't. She can't run. Normally, inside her apartment, Heather doesn't run. No doubt when Heather takes Emily to the park, Heather is pushing the stroller or walking along with Emily, not running around. In Emily's world, running is wrong. It simply isn't done! In fact, it is not only highly unlikely; it is impossible! Yet there Heather is, doing the impossible, the absurd, the intensely wrong. And Emily loves it.
Later the same week I saw Emily with her father, John. I asked him what else makes Emily laughed, aside from running. He said she laughs when people jump. So I tried jumping around. Sure enough: hysterics from Emily.
Think about other things that make toddlers and little kids laugh. Tom recently commented that few things are as funny to his boys, 7 and 9, than their father falling down. Kids often laugh when someone falls or trips. Whereas we often take this as a sign that the child needs correction, or is insensitive to the plight of others, or actually enjoys witnessing pain, maybe something very different is going on. Hypothesis: What they are reacting to is the wrongness of the scene. Laughter places a special flag next to what is wrong, or absurd, or impossible, to make sure that they examine it more thoroughly later. And more: laughter jolts the entire system, and even forces reexamination of existing connections and "established" disparities for the express purpose of learning.
Things that are wrong have no place in the big database--this would be the operational definition of "what is wrong", on my emerging view. Maybe we use laughter to shake it up.
Laughter is a strongly physical event. It focuses the attention on the cause of the laughter; this is precisely why it is commonly thought of as interruptive, disruptive, or distracting. It distracts the attention from other things, or from moving on to other things. But it's not the case that attention or concentration are lost. On the contrary, attention and concentration, on the cause of the laughter, are heightened drastically. It is perhaps a merely Puritanical misconception that laughter and humor are problematic on the whole and indicate a loss or absence of seriousness.
It's worth reminding ourselves at this point that humor is a device that good teachers and speakers use to get a point across and help people remember material.
If Contrived Laughter Causes Pleasure...If contrived laughter is a way of stimulating ourselves to feel just a smidgeon of the pleasure that genuine laughter causes, then that would help explain why we do it. For example, if someone thumps me on the back, I feel a small bit of pleasure; I can simulate a similar feeling by jerking my torso via internal musculature. It's perhaps not the impact per se, not the touch of the other person's hand, but the chain reaction of moving organs inside my body, that cause the feeling. (I'm not talking about the psychological satisfaction of being touched by another, but the raw physical stimulation.) Similarly, in genuine laughter the body convulses; perhaps it is this that stimulates the pleasure centers in the brain, not the mere thinking about something funny. So if we simulate genuine laughter by voluntarily convulsing just a bit, perhaps we can get a little of what we need. Social functions would then be built up upon this self-stimulation. More evidence that this could be the origin of contrived laughter is that sexual satisfaction tends to be more intense (and in some cases more likely or more possible) if one allows oneself to make sounds rather than if one tries to remain religiously and properly silent. Relaxation via meditation and yoga is facilitated by vocalization--this is why there are accompanying mantra sounds that reverberate deep inside one's body. It is something that we can do to ourselves. It may happen more often than we are aware or would like to know.
To be continued.
view entire entry:
2002_03_02:23: Trauma Representation
to see pictures, view entire journal
Because he knows I've been thinking a lot about trauma this year, Frank forwarded a news report about how trauma affects children's brains. And because I'd just been thinking about how trauma might be represented in an artificial intelligence, I responded thus:
Fri, 8 Feb 2002 02:12:09 -0800 (PST)Then, in response to Frank's followup, I expanded on the above trauma thoughts, and connected them to thoughts on dreams:
From email@example.com Wed Feb 27 20:51:06 2002Note: The names of persons in
the above examples have been changed to protect their privacy.
view entire entry:
2008_06_08:14: learning, intelligence, wrongness
to see pictures, view entire journal
One toy problem I worked on to great success recently was what I call "spelling forgiveness". The idea is that, if you give a literate human being a paragraph full of spelling errors, she will still be able to read it, and in many cases may not even notice that there are errors. How is this possible? What is it about the intelligent mind that would enable it to ignore fairly large differences in the configuration of an observable object and recognize it?
One thing I know for sure is that I don't have a list of misspellings in my mind with pointers to the correct spellings. What if someone misspells a word in a way in which I have never seen it misspelled before? How do I read that, if I don't already have that error memorized? I also don't have a list of possible transformations in my mind (ie for ei, rt for tr, ou for uo, ee for ei, that sort of thing); I've never had any affinity for memorizing such things, and again, people can transpose letters in all sorts of unexpected ways. When spellcheckers rely on these methods, they have very limited success.
I was inspired to work on this problem for three main reasons: One, I know a very bright teenager who is so severely dislexic and disgraphic that he spells almost every word in his home assignments incorrectly; he doesn't usually misspell them the same way more than once or twice, and there is no discernible pattern to his mistakes: sometimes he leaves out all the vowels, sometimes only some vowels, sometimes he puts extra letters in, sometimes he spells them phonetically and sometimes not. The upshot is great frustration, because there is no spell-checker that can even guess at what he means. However, his parents and teachers can read the paragraphs. So obviously it is not impossible.
Two, there are games (Scrabble, Boggle, wordfind puzzles) where all the correct letters are there but there may be additional letters confusing the mix and whatever letters are there may be scrambled. Add to this the Cambridge study that shows that people have no problem reading paragraphs wherein the first and last letters of the words are in the correct place, but the internal letters are not; here is a paragraph for your convenience:
Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe. (from http://www.devtek.org/test/cambridge_exercise.php)
See criticisms of this example, same url, that say that there are groups of letters within the word that remain in together, but I think this is simply an accident. If you scramble them up quite thoroughly, people can still read the words, if with more difficulty; and Silhouette can still read them. Further, I don't consider it a criticism of the experiment or the interpretation, but a clue as to how we do this: we do it through recognition, not through incremental searching or by means of the "edit distance" algorithm (always the computer scientist's kneejerk reactions!).
And, finally, three, there are implications for all perception, all knowledge, all thinking. I use words in the toy example, because it is a way to approach and understand the problem that is simple as well as interesting. But this is not about spelling! Spelling words is merely one example of the animal ability to recognize that which is not exactly like what has been seen before. If you see an accident on the road, with parts of a car strewn about, you can recognize that it was a car that was involved in the accident, not an elephant or an airplane. How do you do this? This is the interesting question; questions about language and spelling are, frankly, simply irrelevant, as far as spelling-forgiveness in an intelligent being are concerned.
It is difficult to talk to people about AI. No matter what I tell them Silhouette does, it is not interesting to them. They want to know why she doesn't do this other thing instead. People who are invited to interact with her typically try to "fool" her. For example, if I tell people that she can "forgive" just about any spelling error, they give her made up words that she couldn't possibly know. One wonders whether they thought I had said, "Silhouette can read your mind!" I might even warn the person that she learns language by reading alone and cannot yet hear at all; yet one of their first efforts to fool her will come in the form of phonetic misspellings. (Interestingly, she can get a great number of phonetic misspellings even without the advantage of hearing; but of course the testers will always focus on the ones that they think she won't be able to get without hearing. Why?) I told someone yesterday about the spelling forgiveness, and he wanted to know how she would distinguish typoes from spelling errors. I said that they are essentially the same thing. He said that they weren't, and he gave the example 'commuter' vs 'computer'--how does she know which was meant? I noted that context would be required for this, and she doesn't have that capacity yet. He was not deterred: surely there should be some probabilistic algorithm for determining which was correct. I pointed out that, on the basis of a single word, he couldn't tell which word was meant either. Somehow this answer did not seem to satisfy him.
I am sure that this reaction tells us something important about animal intelligence, but I'm not sure yet what it is. Eric thinks that it is due to my using the now-deprecated expression 'artificial intelligence', which has fallen out of fashion in part because it conjures up such high and unreasonable expectation in people's minds in a world where a Microsoft spell-checker still can't help my friend with his highschool homework. But I think these incidents happen all over the place, and may be nothing more than the exemplification of the principle that one should always take the path of least resistence and say the first thing that comes to mind. Or perhaps they simply have the urge, in the case of AI, to demonstrate that they are in fact much smarter than this stupid machine, so there, ha! It's like competing with a two-year-old.