Transcript for:
Lecture on The Emotion Machine and Society of Mind

so what I'm going to do in this course is discuss mostly ideas that are already in the book called the emotion machine I'm sorry I used that title U and the older book called The Society of Mind which are the books are not quite the same they overlap a bit in material but they're sort of complimentary I like the old one better because the chapters are all one page long and they're moderately independent so if you don't like one you can skip it uh the new book is much denser and it has a smaller number of long chapters and I think it's um over the years I got lots of reactions from um young people in high school for example uh almost all of whom uh liked the Society of Mind and found it easy to read and seem to understand it U there are lots of criticisms by older people who uh maybe some of them found it harder to put so many fragments together who knows but most of this class most of the things I'd like to say are in those books so it's really like a big seminar and I'll uh My Hope Is that everyone who comes to this class would have a couple of questions that they'd like uh to discuss and if I can't answer them maybe some others if you can so I'd like to think of this as a super seminar and normally I don't prepare lectures and uh I just start off asking if there are any questions and if they're not I get really pissed off because um but anyway I'm going to start with a series of slides so why do we need machines and partly there are a lot of problems unlike most species uh or kinds of animals humans have only been around a few million years and uh they're very clever compared to other animals but it's not clear how long long they will last and when we go we might take all the others with us so they're a whole set of serious problems that are arising because there are so many humans and here's just a little list of things there's a better list in a book by the astronomer Royal Martin lease uh of England anybody know the title our yes our final hour it's a slightly scary title and when I was uh te teenager World War II came to an end with the dropping of two Atomic oh this is getting terrible two atomic bombs on Japan and I didn't believe the first one was real because it was in hoshima so I assumed that the US had somehow made a big underground underwater tanker with 20,000 tons of TNT and some few grams of radium or something and blowing it up in the harbor and first it it flew an airplane over dropping some little thing and this was to fool the Japanese into thinking that we have an atomic bomb um but uh when they did it again over Nagasaki that wasn't feasible so and when I was in grade school sometimes if I said something very bright I would hear a teacher saying maybe he's another J Robert Oppenheimer because uh that was the name of a scientist who had been head of the Manhattan Project and he was I think three or four years earlier in in uh grade school than I was and I thought it was very strange for a person to have a first name is just being a letter rather than a name and uh many years later when I was at Princeton in graduate school I met the Robert Oppenheimer and that was a great pleasure and in fact he took me to lunch with a couple of other people I admired namely girdle and Einstein which was very exciting except I couldn't understand Einstein because I wasn't used to people with a strong German accent but I understood girdle just fine and after that lunch was over I went and spent about a year learning about touring machines and trying to prove theorems about them and so forth so anyway um in the course of these talks we'll run across a few of these people and here's a big list of the people that I'm mostly indebted to for the ideas in the Society of Mind and the emotion machine the ones in blue are people I've actually met it would be nice to have met Aristotle because uh no one really knows much about him but uh you really should read just skim through some of that and you'll find that uh this is a really smart guy we don't know if he wrote this stuff or if it were compiled by his students like a lot of fineman's writing is and V nman for Norman's writing is edited from notes by their students and uh anyway the astonishing thing about Aristotle is that he seems to be slightly more imaginative than most uh cognitive scientists you'll run into in the present day it would have been nice to know Spinosa and Kant and the others also Freud wrote 30 or 40 books so did he fall off this list there he is I just made this list the other day and I was looking up these people to find their birthdays and stuff yes why are there no Eastern philosophers because they're religious as far as I can see well who would you would you say Buddha no I mean just Eastern thinkers that name one maybe I never heard of them who confusious confucious or you know L from China well I only know of them through aphorisms single Proverbs but I don't know that confucious had a theory of thinking do you think he did there a lot of different schools ofs well I've looked at buh theories and they're I don't think they would get a C+ and one problem is that there are cultures there's something about Greek culture because it had science it had experiments somebody has a theory and they say and like epimenides lucre uh somewhere in the Society of mind I think I quoted lucretius about uh translucent objects and he says they're they have the particular appearance because the Rays of light bounce many times before they get to the surface so you can't tell where they started and I don't find an Eastern philosophy theories that say here's what I think and here's a reason why why I've looked at Buddhist stuff and it's it's strange lists of psychological principles every one of which is looks pretty wrong and they make nice uh two-dimensional diagrams but no evidence for any of them so I don't know whether to take it seriously I think a lot ofers having knowled from observation that some some of thema didn't really test it because a lot of theology cannot be tested on the other hand are scientists but what can't be tested I mean some of the ideologies probably then why if they can't be tested why should one look at it twice I don't know like culture can you test culture okay I think this is a serious argument it seems to me that science began a little bit in China a little bit in India uh in the Arabic World they got up to the middle of high school algebra but then what that's found well but it wasn't as good as aredes who got to the beginning of calculus so if you look at most cultures they never got to the critical point of getting theories doing experiments discussing them and then throwing them out and so if you look at Buddhist philosophy it's 2500 years old if you look at Greek physics yes Archimedes almost got calculus and he got lots of nice principles and Buddha mentions at some point if you want to weigh an elephant uh put him in a boat and then take the elephant out and put rocks in till the boat sinks to the same level so there you see a good idea but if you look at the history of the culture if people still say this thousand year old stuff is good then you should say no it's not uh there's like a star in Chinese uh history that has the same sure I mean it's I don't think maybe there's no one person no the question is why did it stop why did it stop ancient wisdom is generally not very good and we shouldn't respect it for too long and that's no everybody no we we got rid of alchemy we got rid of uh what do you call it what's caloric you you you you jump off their shoulder you don't stay on them so it's good to know history but if the history doesn't get anywhere then you don't want to admire it too much because you have to ask why did it stop what went wrong and usually it went wrong because barbarians came in and well you know what happened to aredes uh some some Roman killed him but uh anyways no it's it's good it's a good question why why didn't science happen a million years ago because humans are 5 million years old so what took it so long and no it's more sure okay do you have a theory of why science didn't develop for so long in most cultures it might be religion which is a sort of science that doesn't use evidence and in fact kills people who try to get it and so there there systematic reasons why most cultures failed and uh maybe somebody has written is there a book on why science disappeared ex except once it's a rather remarkable isn't it after all the idea if somebody says something and somebody else says okay let's do an experiment to see if that's right you don't have to very be very bright so how come it didn't happen all the time everywhere curious I don't know the answer to that but I know Paul Davies has sort of an anecdote about that where he's he's speculating even in Europe where did happened was it a fluk and I think he gives the example of suppose an asteroid or a comet crash in Paris in you know he 1150 or 1200 or something then what with with sign never developed anywhere he just well history is full of flukes I'm trying to remember who wrote that nice book about the plague some woman and she mentions that uh this was spread by rats and fleas or something and 30 or 40% of the population of many countries in Europe died and uh the Next Generation had a lot of furniture the standard of living went way up so so anyway here's a list of disasters and go come on and Martin Reese is the Royal astronomer and has that book about the the last hour or whatever and uh I'm making another longer list but uh he has lots of obvious disasters like some high school student looks looks up the the genetic sequence for small poox virus has been published and now you can write a list of I mean of nucleotides and send it somewhere and they'll make it for about 50 cents or a dollar per nucleotide for so for a couple of hundred dollars you can make a virus or a few hundred and so one possibility is that some high school student makes some small pox only gets it wrong and it kills everyone so there lots of lots of disasters like that and no one knows what to do about that because uh the the DNA synthesis Machinery are is becoming less and less expensive and uh probably the average Rich private high school could afford one um so there are lots of other things that could happen um but one particular one is this graph which I just made up um an interesting fact is that since 1950 when the first antibiotics started to appear as I mentioned I was a a kid in the 1940s and penicillin had just uh hit the stands and there wasn't much of it and there was a researcher uh lived a few blocks from us whose dog had cancer and so uh its father I don't know what you call the owner of a dog uh sneaked some pen Ain out of the lab and uh gave it to the dog who died anyway but he said well nobody's tried penicillin on cancer yet maybe it will work and uh a lot of people were mad at him because he probably cost some human its life but he said he might have saved a billion humans their lives so ethics ethicists are people who give reasons not to do things and I'm not saying they're wrong but it's a funny job so anyway uh since that sort of thing happened and Medicine began to advance people have been living one year longer every 12 so it's uh 60 years since 1950 another so um that's five of those six so they're living six or seven years longer now than they were when I was born and uh somebody mentioned that that curve stopped the last few years uh for other reasons but uh anyway if you extrapolated that uh you find that the lifespan is going to keep increasing how much we don't know um another problem is that you might discover enough about genetics to get rid of most of the serious diseases maybe just 20 or 30 genes are responsible for most deaths right now and if you could fix those which we can't do yet there's no way to change a gene in a person because invading all the cells is pretty m massive intervention but we'll get around that and then it might be that people will suddenly start living 200 or 300 years now at some point the population has to slow down and so you have you can only reach equilibrium with one child per family and probably less than that and so all the work has to be done by two or 300 year olds and let's hope they're good and healthy um so anyway I think it's very important that we get smart robots because we're going to have to stem the population and I hope people will live longer and blah blah blah and so these robots have to be smart enough to replace most people and um how do you make something smart well artificial intelligence is the field whose goal we has been to make machines that do things that we regard as smart or intelligent or whatever you want to call it and the idea of seriously making machines smart has roots that go back to a few Pioneers like Li nits who wrote about automa and U that sort of thing but the idea of a general purpose computer didn't appear till the 1930s and 40s in some sense the first form of the general purpose computer appears in really in the 1920s and 30s with the work of a mathematician who AML post had n Yu who I happen to never meet uh but we had some friends in common and uh he had the idea of production rules and basically rule-based systems and prove various theorems about them uh then Kurt girdle uh showed that if you had something like a computer or a procedure that uh had the right kinds of rules it could compute all sorts of things but there were some things it couldn't compute unsolvable problems and that became an exciting branch of mathematics and the uh the star thinker in that field was Alan Turing who invented a very simple kind of universal general purpose computer instead of a random access memory it just had a tape which it could write on and read and change symbols and we go back and forth and if it's in state X at C symbol y it will print symbol Z over The X and move to the left or right and just a bunch of rules like that where it's enough to make a universal computer and so from about 1936 it was sort of clear to a large mathematical community that these were great things and a couple of general purpose like computers very simple ones were built in the 1930s and more in the 1940s and in the 1950s big companies started to make big computers which were rooms full of equip equipment and but as you know most programs could only do some particular thing and none of them were very smart um whereas a human can handle lots of kinds of situations and if you have one that that you've never seen before there's a good chance you'll think of a new way to deal with that and so forth and so how do you make a machine that doesn't get stuck almost all the time and I like to use the word resourcefulness although I left an R out of that one ug is there a shorter word so here's a good example my favorite example of U situation where a person is born more or less with a dozen different ways of dealing with something and the problem that I imagine that you're dealing with is this my favorite example example is I'm thirsty so I see that glass of water and I do that and get it actually I am on the other hand if I were here I would never in a whole lifetime do this you never walk out a window by mistake it's a we're incredibly reliable so uh how do I know how far it is and that slide shows you 12 different ways that your vision system that's only your vision system has to measure distances so gradients if things are sort of blurry then they must be pretty far away that's sort of on a foggy day outside and U here's the situation if you assume those are both chairs of the same size then you know that this chair is about twice as far away as that although you don't well and you know how far away they are pretty much by the absolute size if you have two guys that work well then if something is less than 30 feet away you can make a pretty good estimate of its distance by focusing both eyes on some feature and your brain can tell how far apart your eyes are looking so that's there's 12 different things it's more than you need uh lots of people are missing half of those uh lots of people have very poor Vision in one eye uh some people cannot fuse stereo images even though both eyes are seem have 2020 Vision uh and in some cases nobody knows why they can't do that I think I once took a test for being a pilot and they wanted to be sure you could do stereo vision which seemed very strange because if you're an airplane and you're less less than 30 or 40 feet away from something it's too then yes you can't use stereo you could use stereo but it's too late so anyway that's interesting I see if you think of an example where a person has even more 12 of these but it's pretty amazing isn't it that's more redundancy this is too hard to read but somehow I found in an Aristotle essay the idea that you should represent things in multiple ways you might describe a house one person might describe a house as a shelter against destruction by wind rain and heat another might describe it as a construction of stones bricks and Timbers but a third possible description would say it was in that form in that material with that purpose so you see there's two different descriptions one is the functional description it's a shelter the second one is a structural description How It's Made and Aristotle says which is the better description and he dismisses the material one or the functional one is not rather the person who combines both in a single statement and then I found a paragraph by Fineman who says every theoretic medical physicist who is any good knows six or s seven different ways to represent exactly the same physics and you know that they're all equivalent and uh but you keep them all in your head hoping that they will give you different ideas for guessing guessing I should put more dots anyway that whole argument is to say that um the interesting thing about people is that they have so many ways to do things and perceive things and think of things and in some cases we even know that there are different parts of the brain that are involved in one aspect or another of constructing those different representations or descriptions if you look at the one of my favorite books weighs about 20 pounds it's the book on the nervous system by candle and Schwarz and the index to that book is quite a lot of pages long and it mentions 400 different structures in the brain so the brain is not like the well I shouldn't make fun of the liver because for all I know the liver has 400 different uh mini processes for doing things but the brain has distinguishable areas that seem to perform several hundred different functions and uh with a microscope that at first they all look pretty much the same but if you look closely you see different slightly different patterns of how the most layers of the cortex of the brain most parts of it have six layers and each has a population of different kinds of cells there're a lot of cross connections up and down and sideways to other they're arrang in Columns of between 400 and a thousand cells and you have a couple of million of those and there are lots of differences between the columns in different areas and we know some of the functions most cases we don't know much about how any of them actually work with the main exception of vision where the functions of the cells in the visual cortex are fairly well understood at low levels so we know how that part of the brain finds the edges and boundaries of of different areas and textures and regions of the visual field but we do not know uh even a little bit about how the brain recognizes something as a chair and an overhead projector and a CRT screen and that sort of thing so the kind of question that I got interested in was how can you have a system which has a very large number of different kinds of computers Each of which by itself might be relatively simple or might not I suppose and how could you put them together into a larger system which could do things like learn language and uh prove theor and convince people to do things that they would never have dreamed of doing five minutes earlier and stuff like that now the first sort of things I was interested in was in fact how to make how to simulate simple kinds of nerve cells because in the 1950s there was about almost a 100 years really more like 50 years of science discovering things about neurons and nerve cells the axons and dendrites that they use to communicate with other neurons so if you go back to 1890 you find a few anatomists discovering some of the functions of or connections of neurons in the brain and you find a few experimental physicists there was no oscilloscope yet but there were very uh High Gain galvanometers which could which could detect pulse pulses going along a nerve fiber and uh by 1900 it was pretty clear that part of the activity in a nerve cell was chemical and part was electrical and by 1920 or 30 with the cathode ray tube appearing mostly because of Television but um it became possible to do a lot of neurophysiology by sticking needles and brains the vacuum tube appears around 1900 or and you can make amplifiers that can see mli volts and then microvolts so in the beginning of the 20th century there was lots of progress by 1950 we knew a lot about the nervous system but we still don't know much about how you learn something uh in the brain it's quite clear that the things called synapses are involved the connections between two neurons become better at conducting nerve impulses under some conditions uh but no one knows how higher level knowledge is represented in the brain yet and the Society of Mind book had a lot of theories about that and in particular there was a theory called K lines knowledge lines or something that um came partly from me and partly from a couple of other researchers named David Waltz and Jordan Pollock and uh that's a sort of nice theory of how neural networks might remember higher level Concepts and for some reason although that's that kind of work is from around 1980 which is 30 years ago it has not hit the Neuroscience community so if you look at the at the emotion machine book or the Society of Mind in Amazon you might run across a review by a a neurologist named Richard restac who says that uh Minsky makes up a lot of Concepts like Kines and micronemes and stuff like that that nobody's ever heard of and there's no evidence for them and he ignores the possibility that it isn't the nerve cells in the brain that are important but the supporting tissues called glea which hold the neurons up and feed them and he goes on for a couple of insane paragraphs it's very interesting because it doesn't occur to him that you can't look for something until you have the idea aide of it and so here's this 30y old idea of Kines and go and ask your favorite neurologist neuroscientist what it is and he said oh I think that's some AI thing but where's the evidence for it what do you suppose is my reaction to that who's supposed to get the evidence so it seems to me that there's a strange field in Neuroscience which is that it doesn't want new ideas unless you've proved them so I try to have conversations with them but get somewhat tired of it anyway but in this course I'm taking the opposite approach which is that uh we don't want a theory of thinking we want a lot of them because probably psychology is not like physics what's the most wonderful thing about physics most wonderful thing is that they have unified theories there wasn't much of a unified theory until Newton and he got these three laws wonderful laws uh one was the gravitational idea that that things bodies attract each other with a force that's the inverse square of the distance between them another is that uh kinetic energy is conserved I forget what the third one oh equal reaction is equal and opposites if you hit some if two things Collide they transfer equal amount of momentum to both there was a little problem up to Newton's time Galileo got some of those ideas and my impression from Reading him is that he has a dim idea that there are two things around there's kinetic energy which is MV and there's whoops momentum is MV and there's kinetic energy which is MV squared and he doesn't have the clear idea that there are two different things here and you can't blame him I would think what you wouldn't think that two two quantities would combine in two different ways to make two important different concepts well that got clear to Newton somehow and Galileo is a bit muddled although he gets almost all the consequences of those things right but he doesn't get the orbits and things to come out well anyway what's happened in artificial intelligence like most Fields is that people said well let's try to understand thinking and psychology and let's let's use physics as our model and so what we want is to get a very small number of universal laws and a lot of psychologists struggled around to do that and then they gradually separated so that there were some psychologists like Bill eses who worked out some very nice mathematical rules for reinforcement-based learning got a simple rule if you designed an experiment right it predicted pretty well how many trials it would take a rat or a pigeon or a dog or whatever to uh learn a certain thing from trial and error and SD's got a set of four or five rules which looked like Newton's laws and if you designed your experiment very care carefully and shielded the animal from noise and everything else which is what a physicist would do for a physics experiment uh the reinforcement theories got some pretty good uh models of how to make a machine learn but they weren't good enough so here's a whole list of things that happened in the early years of cognitive psychology when people were trying to make theories of thinking and they were imitating the physicists by physics Envy to borrow a term of Freud there the idea is can you find a few simple rules that will apply to very broad classes of psychological phenomena and this led to various kinds of projects uh lots of neural network and reinforcement and statistical based methods led to learning machines that were pretty good at Learning in some kinds of situations but and they're becoming very popular but I don't like them because if you have a lot of variables like 50 or 100 then to use a probabilistic analysis you have to think of all combinations of those variables because if two of them are combined in something like a exclusive or manner um you know I I just put the light pen in a pocket it's either in a left pocket or a right pocket can't be both uh that's an exor uh that will cause a lot of trouble to a learning machine and if there are 100 variables there's no way you could decide which of the two to the 100th Boolean combinations of those variables you should think about and so lots of statistical learning systems are good for lots of applications but they won't cut it to solve hard problems where the hypothesis is a little bit complicated and has seven or eight variables with complicated interactions so um most statistical learning people assume that if you get a lot of partial ones then you can look for combinations of ones that have high correlations with with uh the result then you can start combining them and things get better and better however if the uh mathematically if an effect you're looking for it depends on the exclusive or of several variables there's no way to approach that by successive approximations if any one of the variables is missing there won't be any correlation of the Phenomenon with the others anyway that's a long story um but I think it's worth complaining about because almost all young people who start working on artificial intelligence look around and say What's popular statistical learning so I'll do that that's exactly the way to kill yourself scientifically you don't want to get the most popular thing you want to see what am I really good at that's different and what are the chances that that would provide another thing so you well end of end of long speech uh another problem in the last 30 years and I'm I'm sort of as you'll see during my lectures I think a lot of wonderful things happened between 1950 when the idea of AI first got articulated 1950s and then uh the 20 years after that from 1960 to 1980 a lot of early experiments and I'll show you some of them um looked very promising in fact they may be here we go 1961 Jim swagel was a young graduate student here at MIT he was blind uh he had gotten some retinal de degeneration thing in his first or second year of high school he was told that he would lose all his vision and there was no treatment or hope so he learned brail while he could still see and when he got to MIT he was completely blind but there was a nice big parking lot in Technology Square and he would ride a bicycle and people like susman and Winston and whoever was around would yell at him telling him where the next obstacle would be and uh Jim got better and better at that and uh nothing would stop him and he decided he would write a program that oh I wrote a program that would take any formula and find its derivative was really easy because they're just about five rules like if there's a product UV then you compute U times the derivative of v and plus v * you know U DV plus v duu so I wrote a 20 line list program that did all the algebraic expressions and what it would do is put D's in and then at the right place and then it would go back through the expression again and wherever it saw a d it would do the derivative of the thing after that and nothing to it so slagel said well I'll do integrals and we all said well that's very hard nobody knows how to do it and in fact in Providence at the uh home of the American mathematical Society there is a big Library called debate manuscript project which has been collecting all known integrals for a hundred years and when everybody when anybody finds a new integral that they can integrate in closed form they send uh the formulas to the baitman manuscript project and uh some hackers there were have develop ways to index it so if you had an integral and you didn't know how to integrate it U you could look it up that was pretty big I should say that slagel succeeded in uh writing a program that managed to do all of the kinds of integrals that one usually found on uh the first year calculus course at MIT and got an A in those couldn't do word problems and The Uncanny thing is that if it was a problem that usually took a MIT student five or 10 minutes slagel's program would take five or 10 minutes it's running on a IBM 701 with uh 20 millisecond cycle time it's incredibly slow you can type almost that fast uh and 16k of words of memory so there's no significance whatever to the this accident of time it would now take a microsc or so be 100 million time thousand million times faster than a student uh quite remarkable I don't have a slide uh Joel Moses then uh SEL went and graduated Joel Moses was another student who is is he provos now or what what he got tired of it a terrific student and he set up a project called Maxima for project Max symbolic something or algebra and uh got people uh several people all over the country were working on integration and at some point a couple of them Bobby cavaness and forget the other one found a procedure that could in fact integrate everything every algebraic expression that has a can be integrated in closed form I forget the couple of constraints on it and uh that became a widely used system It ultimately got replaced by uh Steven wolfram's Mathematica but Maxima was sort of the world class symbolic mathematician for quite a few years and uh Moses mentioned to me he had read slagel's program thesis and it took him a couple of weeks to understand the two pages of or three pages of lisp that uh slagel had written because um being blind slagel had tried to get the thing into as compact a form as possible but that's symbolic it's too easy here's it more ambitious one which is three years later Dan babro who is now a vice president doing something at Xerox and it solved problems like this the gas consumption of my car is 15 miles per gallon the distance between Boston and New York is 250 miles what is the number of gallons used on a trip between Boston and New York and it Chomps away and solves that it has about a 100 rules it doesn't really know what any of those words mean but uh it thinks that the word is is equals the distance between doesn't care what Boston and New York is it has a format thing which says the distance between two things and uh it never bothers to you see because the phrase Boston and New York occurs twice in the example it just replaces that by some symbol it was fairly remarkable and generally if you had an algebra problem and you told it to barro barro could type something in and it would solve it if you typed it in uh it probably wouldn't but it was you know it had more than half a chance or less about half a chance so it was pretty good and uh if you look at an outof print book I wrote called I compiled called semantic information processing most of babro program is in that so that's 1964 I'll skip winegrad which is perhaps the most interesting program this was a program where you could talk to a robot that I don't have a good picture in the slide but they're they're a bunch of blocks of different colors they're all cubes in the OR rectangular blocks and you can say uh which is the largest block on top of the Big Blue Block and it would answer you and you could say uh put the large red block on top of the small green block and it would do that and um wiegratz program was of course a symbolic one we actually built a robot and uh I guess we built it second our friends at Stanford in built a robot and they imported winegrad program and they had the robot actually performing these uh operations that you told it to do by typing and it was pretty exciting my favorite program in that period was this one because it's so psychological this is called a geometrical analogy test and it's on some IQ tests a is to B as C is to which of the following five and Evans wrote a set of rules which uh were pretty good at this did as well as 16y olds and it picks this one and if you ask it why it says something like I what have a reason that it moves the largest object down or something like that makes up different reasons but so you see in some sense we're going backwards in age because we're going from U calculus to algebra to simple analogies oh there it is that's one where the largest object moves down I don't know why I have two of them these are for another lecture okay so that was a period in which we picked problems that people considered hard because they were mathematical but when you think about it more you see well those math things are just procedures and it's once you know what llas and uh gaus and those mathematicians Newton and people did you can write down systematic procedures for integrating or for U solving simultaneous algebraic constraint equations or things like that and so there's very little to to it so in some sense if you look at the what you're doing in math in high school in in education you're going from hard to easy it's just that people aren't most people aren't very good at obeying really simple rules because it's so hideously boring or something so we gradually started to ask well why can't we make machine means understand everyday things and um the things that everyone regards as common sense and uh people can do so you don't need machines to do them one of my favorite examples is why can you pull something with a string but not push and there's been a lot of publicity recently about that interesting uh program written at I group at IBM called Watson which is good at uh finding facts about sports people and celebrities and politics and and so forth but there's no way it could understand why you could push pull something with a string but not push and I don't know of any programs that that has that concept or way of dealing with it so that's what I got interested in and starting around the maybe the middle 1970s or late 1970s several of us started to uh stop doing the easy stuff and trying to make theories of how you would do the kinds of of things that people are uniquely good at I don't know of animals well I don't know I'm sure a monkey wouldn't try to push anything with a string maybe it does it very quickly and you don't notice and one aspect of Common Sense thinking is going right back to that idea of vision having a do dozen different systems is that whatever a person normally is doing they are probably representing it in several different ways and here's an actual scene of two kids named Julie and Henry who are playing with blocks it's pretty hard to see those blocks and you can think that Julie is thinking seven thoughts I'd like to see a longer list maybe a good essay would be to take a few examples and say what are the most common micro worlds see physical social emotional mental instrumental whatever that is visual tactile spatial she's thinking all these things what if I pulled out that bottom block you can't see the tower very well should I help him or knock his Tower down how would he react I forgot where I left the arch shaped block that was real uh it's somewhere over here but I don't think we could maybe it's that I don't know I remember when it happened she mentioned that she reached around and it wasn't where she thought it was so common sense thinking involves this um in most cases I think several representations I don't know if it's as many as seven or maybe 20 or what but uh that's the kind of thing we want to know how to do okay I think I'll stop and we'll discuss things but in the next lecture I'll talk about a model of how I think thinking Works um what's the difference between us and our ancestors we know we have a larger brain but if you think about it if you took the brain that you already had in say uh TR I remember the name of the little monkey that looks like a squirrel jumps around in trees anybody know what what maybe what's it's a squirrel like thing you wouldn't know it was a monkey till you took a close look maybe lemur I don't I forget I'll have to anyway if you just made the brain bigger then the poor animal would be slower and heavier and would need more food and take longer to reproduce the joke about more difficulty to give birth I don't know if any animal has the problem that humans have a lot of people die got and so on so how did we evolve new ways to think and so forth and the my first book the uh Society of Mind had this theory that uh maybe we evolved in series of higher and higher levels of management structures built on on the earlier ones and this particular picture suggests that I got this idea from Sigman Freud's early theories uh there's been a lot of Freud bashing recently so you can look on the web I forget the authors but there are a couple of books saying that he made up all his data and there's no evidence that he ever cured anyone and that he lied about uh all the data in mentioned in his 30 or 40 books and so forth yes right but the funny part is that if you look at his first major book 1895 called the interpretation of Dreams it sort of outlines this theory that most of thinking is unconscious and it's processes you can't get access to and it has a little bit about sex but that's not a major feature and it's just full of great ideas that the cognitive psychologist finally began to get in the 1960s again and never give credit to Freud so uh he may well have made up his data but uh if you have a very good theory and nobody will listen to you what can you do his friend Rudolph fleece listened to him and there was a uh another paper on how the neurons might be involved in thinking which was also written around 1895 but never got published till 1950 by uh forget who um called project for a scientific psychology and it's full of ideas that if they had been published might have changed everything because anyway what's on your mind who has what would you like to hear about or who has another theory got a question great so earlier you talked a little bit about how we don't really see in Neuroscience all these things like P lines Etc do you think it's because they're just really hard to find or no one's actually looking for well bestex review says he uses vague IL defined terms like Kon and Micron and uh couple of others and frame and so forth they're very well defined defin better I mean when he talks about neurotransmitters it's as though he thinks that chemical has some real significance any chemical would have the same function as any other one provided there's another receptor that causes something to happen in the cell membrane so you don't want to regard acetylcholine or epinephrine as having um mental significance it's just a it's just another pulse but very low resolution and uh yes a neurochemical might affect all the neurons a little bit and raise the average amount of activity uh of some big population of cells and reduce the average activ of some others but that's nothing like thinking that's like saying in order to understand how a Car Works what's the most insulting thing I could say or to understand how a computer works you have to understand the Arsenic and phosphorus and or what's the other one you have to understand these atoms that are what yeah well that's the that's the Matrix so there are these one part in a million impurities and that's what's important about a computer isn't it the fact that the transistor has gain and so forth well no the trouble with the computer is the transistors that's why practically every transistor in a computer is M to another one in opposite phase to form a flip-flop whose properties are exactly the same except one in a quadrillion times in other words everything chemical about a computer is irrelevant and I suspect that almost everything chemical about the brain is unimportant except that it causes it helps to make the columns in the cortex which are complicated Arrangements of several hundred cells work reliably whereas the neuroscientist is looking for the secret in the sodium when a neuron fires the important thing is that that lets the sodium in and the potassium out or vice versa I forget which at 500 molts really quite a colossal event but it has no significant it's only when it's attached to a flip-flop or to Something Like A Kline which has an encoder and decoder uh of a digital sort uh every few microns of its length that you get something functional so the trouble is the poor neuroscientists started out with too much knowledge about the wrong thing the chemistry of the neur on firing is very interesting and complicated and cute and in the case of the electric eel you know what happened there the neuron synapse it got rid of the next neuron and it just in the electric eel you have a bunch of synapses or motor end plates they're called uh in series so instead of half a volt if you have 300 of those you get 15 50 volts I think the electric shock that a electric eel can give you is about 300 volts and uh this can cause you to drown promptly if if you are in the wrong wave when it happens to bump into you I don't know why I'm rambling this way uh you're welcome to study Neuroscience but please try to help them instead of learn from them yeah and they just don't know what a Kine is and that's a paper that's been widely read published in 1980 and resta say IL defined and I guess he couldn't understand it yeah yeah why instead of trying to make the neuroscientist like trying to find in the human why don't we just like as computer scientist pram like lines and and all this is the like why is not is thate into the computer s well there I'm surprised how little has been done there's Mike Travers has a thesis Tony Hearn there are three Master's Theses on Kines and they sort of got them to work to solve some simple problems but uh I'd go further I've never met a neuroscientist who knows the pioneering work of Newell and Simon in the late 1950s so there's something wrong with that Community they're just ignorant it's uh they're proud of it oh well I spent some time learning Neuroscience when I was I once had a great Stroke of Luck when I was a I guess I was a junior at Harvard and there was a great new biology building that was just constructed you probably know it's a great big thing with two rhinoceroses what are those what are those two huge animals so this building was just finished and half occupied because it was made with the future so I wandered over there and I met a professor named John Welsh and I said I'd like to learn neurology and he said great well I have an extra lab why don't you why don't you study the crayfish claw I said great so he gave me this lab which had four rooms and a dark room and a lot of equipment and nobody there and he had worked on crayfish so there was somebody who went every week up to Walen Ponder somewhere and caught crayfish and bring them back and I was a radio amateur hacker at the time so was good at Electronics so I got my crayfish and well showed me how to the great thing about this prep preparation is you can take the crayfish and if you claw and if you hold it just right go snap it comes off grows another one takes a couple of years uh and then there's this white thing hanging out which is the nerve and it turns out it's six nerves one big one and a few little ones and if you keep it in ringer solution whatever that is it can live for several days so I got a lot of switches and little inductors and things and made a gadget and mounted this thing with six wires going to these nerves and then I programmed it to reach down and pick up a pencil like that and wave it around well that's obviously completely trivial and all the neuroscientists came around and gasped and said that's incredible how did you do [Music] that they had never thought of putting the thing back together and making it work anyway it was um I always reminding myself that I'm the luckiest person in in the world because uh every time I wanted to do something I just happened to find the right person and they'd give me a lab I got an idea for a microscope and there was this great Professor purcel who got the Nobel Prize after a while and he said that sounds like it would work oh why don't you take this lab uh it was in the Jefferson anyway yeah part of the reason that you don't see experimental neuroscience and things like Kines is that neurons are long and thin so if you want to do an experiment to actually me real neural network you have to trace structures with roughly maybe tens of nanometer resolution but you need to trace them over what might be a couple or even tens of millimeters to T to we and you need to do this for thousands and thousands of neurons before you could get to the point of seeing something like a Kine understanding it so it's just a massive data acquisition and processing problem well but they're doing that they're starting to track it they don't they don't know what to look for maybe don't maybe you don't have to do so much maybe you just have to do a few sections here and there and say well look there were 400 of these here now there's only 200 it looks like this is the same kind maybe you don't have to do the whole brain no I even tracing even getting a single neuron is because you it might get down to you need to be looking at electron micrographs of grains that are sliced at about 30 mm excuse me 30 NM so even just having a single person reconstruct a single in N takes might take weeks well I don't know maybe a bundle of K lines is uh a half a millimeter thick oh so if you actually do some larger scale structure to start looking at yeah why not I just think they have no idea what to look for I could give you 20 of those in five minutes but who nobody's listening before you could look for it at that scale what scale I don't I mean they know what neurons look like so you know you know what to look for if you're saying the of neuron that level I'm saying you may only have to look at the white matter oh ignore the neurons because the point of Kines is where do these go and uh what goes into them and out I don't know it's just this idea let's map the whole brain a 100 billion things and then people like rest say oh and there's a thousand supporting cells for each neuron he's just glorying in the obscurity of it rather than trying to contribute something anyway if you run into him give him my regards I really wonder how somebody can write something like that yes excuse my ignorance but what is a KINE the idea is that suppose one part of the brain is doing something and it's in some particular State that's very important like I don't know that like uh I've just seen a glass of water then another part of the brain would like to know there's a glass of water in the in the environment and I've been looking for one so I should try to take over and uh do something about that now at the moment there's no theory of what happens in different parts of the brain for a simple thing like that to happen no Theory at all except they use the word association or or they talk about what are the purposeful neurons goal forget okay so my theory is that there there a bunch of things which are massive collections of of neres fibers maybe a few hundred or a few thousand and when the visual system sees an apple it turns on 50 of those wires and when it sees a pair it turns on a different 100 or 50 of those wires but about 20 of them are the same so forth in other words it's like the edge of a punched card have you ever seen a card-based retrieval system if you have a book that has uh suppose it's about physics and biology and uh Sumatra and a typical 5 by8 card has 80 holes in the top Edge so what you do is if it's if it's Sumatra you punch eight of these holes at random particular set they're assigned to the Sumatra and then if it's I forget what my first two examples were but you punch eight or 10 holes for each of the other two words so now there are 24 punches only probably four or five of them are duplicates so you're punching about 20 holes and now if something is looking for the cards that have were punched for those three things even if there are 30 or 40 other holes punched in the cart you stick your 20 wires through the the whole deck and lift it up and only cards fall out that had those three categories punched for so you see even though you had 80 holes you could punch combinations of up to a million different categories into that and uh if you have to put a bunch of wires through you'll get all of the ones that were punched for those Cate the categories you're looking for and you might get three or four other cards that will come down also because all of the eight holes were punched for for some category by accident do you get the picture it's I'll send you a reference it was invented by a uh in the 19 early 1940s by a uh Cambridge scientist here named Calvin Moors and was widely used in libraries for information retrieval until computer came along but anyway um that's the sort of thing you could look for in a brain If you had the concept in your head of zetoc coding but I've never met a neuroscientist who ever heard of such a thing so you have this whole Community which doesn't have a set of very clear ideas about different ways that knowledge or symbols could be represented in neural activity so good luck to them when they get their big map uh they'll still have to say what do I do with a 100 billion of these interconnections yeah what are your thoughts about the current artificial intelligence research at MIT such as Winston's project I think Winston is just about one of the best ones in the whole world I don't know any other projects that are trying to do things on that higher level of uh Common Sense knowledge he's just lost a lot of funding so one problem is how do you support a project like that have you followed it I don't know if there's a recent summary of what what they're we used to write a big a new book every year called the progress report the nice thing is that we never wrote we had a very good form of support from arpa or DARPA uh which was every year we every year we'd tell them what we had done they didn't they didn't want to hear what we wanted to do and things have turned the opposite so what would happen is every year we'd say we did these great things and uh we might do some more went on for about 20 years and it was and then it fell apart one thing it's a nice story there was a great liberal Senator Mike Mansfield and unfortunately he got the idea that the defense department was getting too big and influential so he got Congress to pass a law that ARA shouldn't be allowed to support anything that didn't have direct military application and Congress went for this and all of a sudden uh a lot of research disappeared basic research it didn't bother us much because we made up applications and said well this will make a military robot that will go out and uh do something bad I don't remember ever writing anything at all because uh but anyway around 19 1980 uh the um the the funding for that sort of thing just dried up because of this political accident it was just an accident that arpa mainly through the Office of Naval Research was funding basic research and that's that was a bit of History uh if you look back at uh the year 1900 or so you see people like Einstein making these nice theories but Einstein wasn't a very abstract mathematician so he had a mathematician named Hermon v um polishing his tensors and things for him and Herman vile's son Joe was at the office of research in my early time and uh that office had spent a lot of secret Money Getting scientists out of e Europe while Hitler was marching around and sending them to places like Princeton and other forms of Heaven in the in Cambridge and again one of the reasons I was lucky is that uh I was here and all these you know if you had a mathematical question you could find the best mathematician in the world down the block somewhere and uh Joe vile was partly responsible for that and the onr uh was piping all that money to us for work on early AI so it's a very sad thing of the maybe the most influential liberal in the US government actually ruined everything by accident arpa changed its name to DARPA it was Advanced research projects agency and it had to call itself defense Advanced research project agency yeah Prof think the achievement of artificial intelligence is inev or is well Christianity wiped out science that might happen tomorrow only uh only choose your religion favorable circum it's a hard problem U the number of people working on Advanced ideas in AI is has gotten smaller and smaller as the right now the around 1980 rule-based systems became popular there were lots of things to do uh right now statistical based inference systems are becoming popular and as I said these things are tremendously useful but the problem is if you have a statistical system the important part is guessing what are the plausible hypotheses and then making up the then finding out how many instances of that are correlated with such and such so it's a nice idea but but the hard problem is this is the abstract symbolic problem of what sets of variables are worth considering at all when there are a lot of them so to me the U the most exciting projects are the kind that Winston is developing for reasoning about real life situations and the one that Henry Lieberman would you stand up Henry liberman runs a worldclass group that's working on Common Sense knowledge and uh informal reasoning and it seems to me that that's the critical thing that all the other systems will need uh in the meantime there are people working on logical inference which has the same problem that statistical inference has namely how do you guess which combinations of variables are worth thinking about then it seems to me that the statistics isn't so important in fact there's a great researcher named Douglas lenat in Austin Texas who once made an interesting AI system that uh was good at uh making predictions and guessing explanations for things and it was sort of like a probabilistic system it had a lot of hypotheses and every time one of them was useful in solving a problem it moved it up one on the list so lenet thing never used any numbers it didn't say this is successful 73 of the time and now it's successful 736 4825 of the time what it would do is if something was useful it would move it up past another hypothesis every now and then we put a new one in well if you're doing if you're trying to solve a problem what do you need to know you want to know what's the most use what's the most likely to be useful one and try that you don't care How likely it is to be useful as long as it's the most right I mean if it's one in a million maybe you should say I'm getting out of here I don't I shouldn't be working in this field at all or uh uh get a better problem but Len thing did rather wonderfully at making theories by just changing the ranking of the hypothesis that it was considered no numbers it it did something very cute uh he gave it examples of arithmetic and it actually it was rather long effort and it actually learned to do some arithmetic and it invented the idea of division and the idea of prime number which was some number that wasn't divisible by anything it decided that nine was a prime didn't do much harm and uh it crept along and it got better and better and uh it invented modular arithmetic by accident at some point and uh it's a PhD thesis a lot of people didn't believe this PhD thesis because lenette lost the program tape so uh so uh he was under some cloud of Suspicion for people thinking he might have faked it but who cares anyway uh I think there's a lesson there which is that uh let's start with something that works and then uh if it's really good then hire a mathematician who might be able to optimize it a little but the important thing was the order and the a good statistical one might waste a lot of time because here's this one that's 78 and here's this one that's 56 and it's the next one down and you get a lot of experience and it goes up to 5'7 and 5'8 and it never you know might be a long time before it gets past the other one because you're doing arithmetic whereas in lennet it would just pop up past the other one then it would get tried right away and if it were no good it would get knocked down again who cares so it's a real question of I don't know is mathematics is great and I love it and a lot of you do but there should be a name for when it's actually slowing you down and wasting your time because there's a better way that's not formal yeah they were saying there are people who know the price of everything and the value of nothing yes that's very nice yeah um I know you're also a musician um so I have like a music related question what do you think is the role of music like why do all cultures happen I have a paper about it oh okay um I've been trying to revise it actually but it's a strange question question because there is music everywhere uh on the other hand I have several friends who are who are a musical and so when I have this theory that uh music is a way of teaching you to represent things in an orderly fashion and stuff like that um well I have three of my colleagues who aren't musical but they dance so they may it may be that I don't know the answer it's interesting the the theory the first theory in my paper is that when you have a lot of complicated things happening then the only way to learn is to represent things that happen and then look at the differences between things that are similar and then try to explain the differences right I mean what else is there maybe there's something else so in order to become intelligent and understand things you have to be able to compare things and to me the most important feature of what what's called music is that it's divided into measures and measures are the same number of Beats or whatever they are and so now you can say what's the difference you change the eighth notes in the second one the last four eighth notes no the two before last to a quarter note so uh you're taking things that were in different times and you're superimposing those times and now you can see the difference and the reason you can see the difference is that uh you have things called measures and the measures have things called beats and so things get knocked into very good frames now there's some Indian music which has 14 measures for a phrase and some of the measures go seven and five and I can make no sense of that stuff whatever and I've tried fairly hard but not very so I don't understand how Indians can think any of you can handle Indian music I guess just to add on what uh you said about um this myor quote from your paper on music mind and meaning is the one about uh what good is music about how kids play with blocks to learn about space and um people play with music to learn about time and I think um in that sense both music and dance are different ways that people can arrange things in time and in a sense like improvisatory music and proposito movement are both ways of of like different blocks if you will time as opposed to space mhm yeah my friends who seem a musical they probably maybe there's something different about their CA or maybe they have absolute pitch in some sense which is a bad thing to have because if you're if you're listening to a piece composed by a composer who doesn't have absolute pitch then you're reading all sorts of things into the music that shouldn't be there and if you're and the opposite would be true I read music criticism sometimes and maybe the reviewer says and after the second and third movement he finally returns to the initial key of E flat major what a relief well I once had absolute pitch for a couple of weeks uh because I ran a tuning fork in my room for a month and uh and I didn't like it because you can't you can't listen to Bach anymore oh well it's a good question H why do people like music and um I don't know any other paper like mine if you ever find and I'd like to see it because if you go to a big Library there are thousands of books about music and if you open one it's mostly berlios complaining that somebody wouldn't give him enough money to hire a big enough chorus but I found very few books about music itself yes anymore is that all right um do you think that having a body is a necessary component of having a mind then I mean could you you do just as well just as a you know sort of a simulated creature oh sure you have all the things simulation I think a mind that's not in a world wouldn't have much to do it would have to invent the world and uh I don't see why it couldn't but you might have to give it a start like the idea of three or four dimensions but can't you what happens if you sit back and just think for a while uh you wouldn't know if your body had disappeared for for would you there also a strange ideas about existence and why do you think there's a world one of the things that bugs me is people say well who created it and that can't make any sense because this is just a possible world suppose there are a whole lot of possible worlds and there's one real one how could you ever how could you possibly know which one you're in and then you could say well didn't someone have to make it and what's the next thing you'd ask well who made the maker so the body mind thing seems to me that uh once you have a computer uh it's it can be its own world it just can the program can spend half the time simulating a world and half the time uh thinking about what it's like to be in it yeah uh yes it's an empty concept it's all right to say this bottle exists because that's saying this bottle is in this universe but what would it what would it mean to say the universe exists the universe is in the universe so so there's something wrong with thinking about so they're only possible worlds there's no it doesn't make any sense to pick one of them out and say that's the real one yeah but it's existence is relative yeah you don't say that's this is this is the world I'm in but you shouldn't say it that doesn't mean it exists like two is in the set of even numbers but what's the set of even numbers in it doesn't stop anywhere yeah there lots of worlds well you can't tell because five minutes from now everything might change so pH nothing ever explains anything you just have to take what you've got and make the best of it yeah Rel which knowledge well there are people who talk about systems theory but I'm not sure that it's well defined artificial intelligence means to me uh making a system that is resourceful and doesn't get stuck and so if you have a system and also it's a how do you put it uh some definitions are not stationary like uh what's popular popular is what's popular now there there isn't any such thing as popular music in terms of the music so I know there were there was once a little Department called systems analysis at tfts which had a couple of rather good philosophers trying to make general theory of everything and they were writing nice little papers and it got it moved along but then there was a Senator McCarthy you've probably heard of and he announced that he had evidence that the one of the principal investigators had slept with his wife before they were married well T was very frightened at this and abolished that department and bill shutz went to California and started esselin and had a good time for the next 50 years I don't more stories yeah kind of as a extension of the body and mind question it seems to me like we as humans we learn a lot from just interacting with the environment like language we hear it being spoken we speak it you know we see things we touch things uh but it as far as I know it a lot of the efforts in artificial intelligence so far have been confined to the computer that does not go the real world interact doesn't kind of Sur learn new things um well here's the problem um I look over at Carnegie melon and there are some nice projects and the most popular one is robot socker and here are these little robots kicking a ball around they're Sony what are they called yes the Sony eye bows so only stopped making the eye bows but it respected Carnegie and it made a little stash secret stash of eye bows to send to Carnegie if when the present ones break but my impression of AI projects that have robots is that they do less less less than projects that don't the reason is if you have a robot like Asimo made by Sony no Honda Asimo can get in the back seat of a car with some effort usually Falls over however if you simulate a stick figure in a computer getting into a stick figure of a car then you can make it learn to do that and get better and better and so all AI projects with without robots are way ahead of all AI projects with robots and the profound reason is that robots are usually expensive and they're always being fixed so if you have five students and the robot is being fixed I don't know what they're doing but they have to wait whereas if you have a stick figure robot then uh you can just run it on this although it might be a little slower than your main frame probably not yeah back to the idea of the mind and the body um here's here's a theory that I just thought of um the idea of the body as a seen in abstract basically a mechanism for input output it's a set of sensors um in In from which our brains can get information about the world and a of in which we can display our states um so in given in that light it's it's almost as if our brains aren't really independent on our body itself it can adapt to like any sort of body if we happen to hook it up that way and it's it just so happens that we've been hooked up to this body since bir that we have such like good mental models of how to use this body um and I guess an example of um in from experiments that this Theory might be how um when people have limbs amputated it takes them a while to like forget that they they have the limb because their mental models still exist the mental models don't go away overnight um and also I guess they train monkeys to control robot with their brains sure well but it just seems to me that a large amount of our brain is involved with high highly evolved Locomotion mechan mechanisms and uh as I said when you're sitting back with your eyes closed in a chair thinking about something then it's not clear how much of that Machinery is important but it might be that um I have a I have a strange paper on I don't know if it's trying to remember its name it's called uh do you think I can actually get a I can't remember the name of the title oh I give up um the idea is that maybe in the older theories of psychology everything is learned by experience in the real world so conditioning and reinforcement and so forth uh in this Theory I call internal grounding I make a conjecture suppose the brain has a little piece of nerve tissue which consists of a few neurons arranged to make not a flip-flop but a what would you call a three or a four flop a flip-flop with three or four states let's say three so when you put a certain input it goes from I couldn't find the TRU so here are three states and here's a certain input that means if you're in that state you go to this and if you pop that input again it does this and if you say uh go clockwise go counterclockwise it goes so three of them get you back where you were but if I go this this and that that would mean to go like this this and back so this would be that means that's equivalent to just going one get the idea in other words imagine that there's a little world inside your brain which is very small and only has three states and you have actions that you can perform on it and you have an in eye which can see which of the three points of that triangle you're on then you could learn by experience that if you go left left left you're back where you were but if you go left right left right you're back where you are and if you go left left right that's like going one left in other words you could imagine a brain that starts out before it connects itself to the real world it starts by having the top level of the brain connected to a little internal World which just has three or four states and you get very good at manipulating that then you add more sensory systems to the outer world and you get to get learn ways to get around in the real world so I called that the internal grounding hypothesis and my suggestion is maybe somewhere in the human brain there's a little structure that's somewhat like that which is used by the frontal part of the cortex to make very abstract ideas you understand the more abstracted idea is the simpler and more stupid and Elementary it is abstract doesn't mean hard abstract means stupid real real things like this are infinitely complicated so we might have and I wouldn't dare suggest this to a neuroscientist there might be some little brain Center somewhere near the frontal cortex that allows the frontal cortex to do some uh predicting and planning and induction about uh very simple a few simple finite State Arrangements who knows would you look for it well if you were a neuroscientist you could say oh that's completely different from anything I ever heard let's look for it and if you're wrong you're wasted a year and if you're right then you become the new uh ramoni kahal or someone who's the best new who's the currently best neuroscientist maybe it's late one more question one last question uh this is Cynthia Salman who's one of the great developers of the logo language yay yes maybe it's bad question for the end what do you think about theories such as Rodney Brook's theories that speak of completely weird obviously those theories have nothing to do with human thinking but they're very good for making stupid robots and the vacuum cleaner is one of the great achievements of the century it however his his projects what was it called Cog disappeared Without a Trace that theory was so wrong that it got a national award and it corrupted AI research in Japan for several years I can't understand Brooks became popular because he said maybe the important things about thinking is that there's no internal representation you're just reacting to situations and you have a big library of how to react to each situation well David Hume had that idea and he was a popular philosopher for hundreds of years but it went nowhere and it's gone and so is Rod however he is one of the great robot designers and he may be the instrumental in fixing the great Japanese nuclear meltdown because they're shipping some of his robots out there the problem is can it open the door so far no robot can open the door even though it's not locked I usually start by asking if there are any questions but I thought I'd say a few things about chapter one and then see if there are any questions I can't see the pointer oh anybody remember how to get word to make its pointer not disappear uh maybe I mentioned this the first lecture but uh I was taken by this uh cute poem by darthy Parker uh because the first chapter was about love and stuff like that um so I tried to get the rights to reproduce it and it turned out that uh she was angry at all her friends she must have been a perpetually pissed off person and so she left all her literary rights to the NAACP and I called them up for hours and they couldn't find the rights so finally so it's in the uh version of the emotion machine on the web but uh I had to resort to Shakespeare to replace her Shakespeare's a slightly better poet but he's not as funny as daroy Parker so the first chapter starts out or it's mostly about all the things we don't understand about the mind which is almost everything and the first discussion is well the whole chapter is making fun of the most popular ideas and the most popular idea of the mind is that people think that um that they're not doing the thinking but there's something inside them that's doing the thinking and it's a uh this idea that there is a self is uh embedded in just about everything we say and think think and really it's hard to see how you would do without it but if you ask what is the self then since this idea is so popular people begin to believe that there is such a thing and it takes all sorts of various forms and the the most dangerous form maybe is the one that religions uh exploit which is that in inside a person with all their complications there's a little pure Essence called the soul or the self or whatever you want to call it and uh it's impossible to describe it or explain it in physical terms um and so that is one of the reasons why we think there are two worlds a physical world and uh lots of other kinds of Worlds each of us has some imaginary model of what what they are and what they're in and and philosophers talk about it and existentialists and so on so there are lots of problems about our ideas about ourselves and in Reading around for half my life uh I was puzzled at the strange ideas that that are around and in Aristotle uh I find the first intelligible theories of mind and emotions so if you look at particular Aristotle's there are number of books and one of them is called rhetoric and it's full of theories about how people reason and influence each other and uh um I'll show you some quotes from that because when I look at the history that I've encountered about psychology uh sort of Aristotle stands out as being the first and and among the best and as far as I can see there were no psychologists nearly as good as him of course we don't know whether there was a hymn exactly uh because he uh what we have of Aristotle and there's a lot of writing but it's all cobbled together by students from all sorts of manuscripts by other people and people who took notes and U Aristotle claims to have learned a lot from Plato well we have very little writing from him and so there you go three centuries before the Christian era as it's called and then uh a couple of thousand years later you start to find people like Spinosa and Kant and uh John Lock and David Hume Who start to make psychology theories very little of which is as good as the ones that uh Aristotle has in all his fragments so one question that frequently bothers me and it should bother everyone is uh why did science disappear for a thousand years and the standard explanation is the rise of the great religions um and why did it come back and you see with uh the first signs of anything like modern science at least in my view with Galileo and Newton there a couple of people before that uh there's some people in the Muslim world uh who invented some high school algebra and they make a big fuss about that it looks like Archimedes uh in a very recently discovered manuscript uh computed in integral he found the volume of a cone which is what is it one6 BH I forget uh anyway uh so why did science disappear and why did psychology appear so late because there isn't much psychology in the modern sense until 1900 or the late 1800s with Francis gton and uh William James lived around around here and um Rudolph fce who Sigman Freud starts writing in 1895 uh people make fun of Freud as I mentioned last week uh but uh in fact among other things how many of you have read the recent criticisms of Freud which claimed that he was a complete Faker and never cured a single patient this is popular stuff I don't believe Freud ever really claimed to cure a single patient so the critics who are really very ferocious claim that he made up all his data and so forth but most of what Freud says is that psychoanalysis is might be a good way to find out what you're really thinking and discover more about yourself and your goals and so forth and uh I had the good or bad luck uh to be introduced to elron hubard by John uh when I was an undergraduate John Campbell was the great editor of the I think it was called astounding science fiction in those days what a marvelous title and uh this uh fairly mediocre science fiction writer elron hubard invented a new form of uh Psychiatry called what was it called DS dianic it's pretty good and uh I'll tell you that story another time but uh John Campbell had Thanksgiving in the commander Hotel every year and invited a bunch of friends and I don't remember if that's how I got to meet Asim of and hind Line and other people but anyway I did and science fiction had a big influence on me from my actually early years but uh starting in college it got very serious anyway uh so chapter 1 starts to talk about the this phenomenon of Psy ology and one of the funny parts is this little section three 1.3 of uh trying to say what are emotions and I looked up emotions in dictionaries and uh can you all read that I don't feel like reading aloud uh but uh there's lots of discussion of emotions and how mysterious and complex they are and then The Marvelous thing is how many words there are for emotional states uh I think I got 300 but I don't remember anyway here's from a to d and I don't recall how I found those but I think but that's a lot how many words for ways to think are there now that's a serious question because I complained maybe on the next page no I didn't uh I found myself complaining that there were very few words for ways to think and then this afternoon when I was pruning these slides it occurred to me that I didn't really try so uh maybe I just didn't think enough so if there a couple of hundred words for everyday emotions uh if any of you can find me a list of 10 or 20 common words for styles of thinking I'd appreciate it because uh I wonder if there are a lot and if not why not so here's a list of typical situations grieving for a lost child Panic at being in an enclosed space I'm not sure any of the words in the list of 300 standard emotions are good enough to describe how you feel for any of these uh Not Unusual states have you ever lost control of your car at high speed no but when I first learned to drive I couldn't believe that you could read signs at the same time as well anyway one of the very best psychologists in history or pair of psychologists uh aren't even called psychologist these are two guys named uh Conrad Lauren and Nico tinbergen and they somebody made up the word ethology what they study is the behavior of animals and in some sense uh presumably they're studying the psychology of animals because uh just as with the person uh when somebody flies into a rage you're not describing their mental state you're describing something about how they behave and so the ethologists too are psychologists and Tin Bergen and Laurence uh starting around 1920s started to analyze the behavior of animals in great detail and uh so here's an example of how a certain fish behaves I actually forgot which fish it is but there's a picture of it some in the book and uh at different points in its in its life it's in different phases and this is just one diagram of a dozen for this for this particular fish and it's reproduction which involves an environment with plants and other things and he divides Its Behavior into parenting courtship nesting and fighting and uh then you see each of those has a lot of subdivisions and uh tinbergen and lorence and some students discovered all these things by sitting in front of fish tanks and watching the fish for uh months and years uh tinbergen also uh spent years on some beach watching seagulls and so he has a diagram like this for uh particular class of seagulls uh when I came to Boston uh my friends and I used to go to Nahant and look at these tide pools there where there are a lot of activities and it was very interesting and I got a big fish tank and imported all sorts of little animals and plants from uh the tide pools in the haunt and I watched them for about a year and didn't learn anything and uh uh that was before I read tinbergen and then I realized there's something about those people which is they could watch a fish and recognize all sorts of behaviors and I would just watch a fish and wonder whether it was hungry or wouldn't you get bored swimming back and forth for three years in us anyway so here the great psychologists of the of our day Aristotle 2,000 years ago and Laurence and tinbergen in the 1930s and Sigman Freud and uh gton and uh William James around 1900 and then what went wrong there's almost no good psychology between then and 1950 50s when something called cognitive psychology started and it was partly due to people who said let's make science let's make psychology more scientific and you've probably all heard of Pavlov or Watson and what happened is around 1900 some psychologist said well these gon's and freuds and William James are very poetic and expressive and literary and they write much better than we do and tell good stories but they're not scientists and they don't do reproducible experiments so what we have to do is simplify the situation to find the basic laws of behavior so let's let's take a pigeon and put it in a vacuum in the dark well well well they didn't go that far but they did put it in the dark and uh there were were two illuminated levers to work and you could make a sound and um the sound could be very annoying or you could have a uh bright annoying flashing light or something and the animal would push one of two levers and one of them would make the stimulus even more annoying and one of them would make it go away and you'd plot curves of how often the animal pressed these levers so you would get a quantitative theory of how much it learned and how much it could remember and how many trials it would have to do and um then uh instead of just looking at reactions to stimuli they uh quickly switch to trying to teach the animal things by giving it two two Alternatives turn left or turn right or push this button or that or whatever and if they push the one you approved of then you'd give them a little pellet of food and there was a lot of engineering so that you would make sure that the food got to them right away because if there were a 10-second delay between an action and a reward the pigeon or squirrel or rat or cat or dog would learn much less quickly than if there was a one second delay and U anyway that went on for 50 years starting around 1900 Pavlov and his dogs and there's a great movie that some guy came around with that had been taken of Pavlov's lab and it shows the sort of like a Great Dictator or something there's this room with lot of cages and dogs and mostly dogs in this case and Pavlov comes in and there are a bunch of Lackey who sort of bow and scrape because he's a Lord and he comes in and all the dogs run into the corner of their cage and and Yelp so uh the uh pavlovian tried pretty hard to get that movie suppressed and I haven't seen it in recent years but uh but anyway uh Fred Skinner who was a professor when I was under graduated Harvard uh was the first one to really automate this uh experiment of psychology and he invented what's called a Skinner box but it's just a soundproof lightproof well ventilated and uh thermally regulated cage and you can put a rat or a pigeon those are the most common animals they're very inexpensive because they're free uh no one knows much about [Laughter] dolphins they've been studied for 50 years whenever John what's his name remember the name of the great dolphin Lily thanks um he discovered a lot about dolphins and a certain amount about their communication uh and a little bit about whales but there's an interesting mystery uh I forget which whales but uh some whales have a 20 minute song and they repeat it for a whole season and next year uh that song is a little bit different but it goes essentially without repeating it's very complicated for 20 minutes and people have studied that a lot and no one has the slightest idea of what it means and nobody even has any good conjectures which bothers me what I think it probably means is this when there's a whole bunch of fish somewhere for uh one of these whales it might be 200 miles away and it's very whales eat a lot and it's very important to find where the fish are and uh I believe this this uh message which changes a bit during the season might be telling you where the food is on the Atlantic or Pacific Coastline in great detail because uh if somebody finds a lot of fish somewhere you have to swim 300 miles and if they're not there U so anyway it's interesting that uh John Lily got a lot of publicity but he didn't discover Squat and finally uh the doll Dophin dolphin studiers gave up because nothing happened anybody have heard anything and I haven't paid any attention for 20 years have you heard of anybody discovering anything about dolphins except they're very good at solving a lot of physical problems anyway that's I'm bothered by the mystery of why was there some psychology in Aristotle's time and Why did Why didn't it get anywhere till 1950 when the there was regular psychology but it was afflicted by what I call physics Envy namely you run into people like es and well he was pretty good actually but there are a lot of psychologists who made up uh things like Maxwell's equations for how animals learn and there were generally three or four laws and uh if there's a sequence of events then animals remember a lot about the first few and the last few in the sequence they don't remember much about the middle and of course the uh reliability of their memory depends a lot on how recent it was and on how powerful the reward was and blah blah and so they get these little sets of rules that look like Newton's laws and that was the kind of sort of psychological physics that the so-called behaviorists were mostly looking for uh this was not what tinbergen and lorence did because they wrote books with extensive descriptions of of what the animals did and made little diagrammatic guesses is about the structure of the sub routines and substructures um anyway end of History uh but it's a nice question why do some Sciences grow and why was psychology the just about the last one I suspect it was it could have been earlier but uh people tried to imitate the physicists and try to say maybe there's something like Newton's or Maxwell's laws for the mind and they found a few but they weren't enough to explain much so there are a lot of questions uh uh when Seymour paper and I started thinking about these things which was really around 1960 um I had been working on some ideas is about AI in the late 50s and my PhD thesis was a a theory of neural networks which was sort of interesting but um never really went anywhere um I went to a meeting in London somewhere and gave a talk about a theory of learning that was based on some neural network ideas and there was this uh person from South Africa named Seymour papert who gave the same paper and uh I hope this happens to you someday find somebody who thinks so much like you only different enough that it's worth it and that uh you only have to say about three words a day and some whole new thing starts because uh we really did write the same paper and it had the same equation in it U and he had been working for p in the who was the first great child psychologist I should have mentioned P who probably discovered more things about psychology than any other s single person in history and uh there are lots of people now who saying he was wrong about 73 uh because children learn that at the age of Two and a Half instead of two and 3/4 I'm parodying the PJ critic Community but it's pretty bad and I think those poor guys are uncomfortable because uh Jean P published 20 books full of observations about children that uh as far as I know no one had made systematically before and uh in his later years he started courting algebraic algebraic mathematicians because he wanted more formal theories and uh in my view he wanted to make his theories worse and uh nothing much happened but he did visit here a couple of times and it was really exciting to meet the starter of a whole new field anyway uh paper and I discussed lots of things and somehow or other we kept finding other more ideas about psychology and it finally jelled into the idea that well if you look at the brain you know that there are several hundred different brain centers what's all that stuff for and uh how could it possibly make any sense to try to uh explain what it does in terms of four laws like Newton like how does a car work is there a magical Force inside the engine that causes the wheels to turn no uh there's this funny thing in the back to cause a differential so that if the car isn't going in a straight line uh the two wheels going at different speeds won't rip the uh if the two wheels were going at the same speed the tread would come off in five minutes do you ever wonder what a differential is for well so most of the car is fixing bugs in the other parts that's most of the brain is because we started out as fish or Lizards or whatever you like and making the brain bigger wouldn't help much because you just get a heavier lizard that had to eat more and would think more slowly so size is bad um but on the other hand if you need another cubic inch of brain to fix the bugs in the other part then you um then the evolutionary advantage of being smarter uh had better make you able to catch a little more food per hour and uh so each person is an Ecology of these different processes and um the brain reached its present size about a million years ago I guess what's what's the current guess anybody been tracked they keep discovering new ancestors of humans and I don't have the patience to read about them because you know that next week somebody will say oh that isn't in the main line and you were just unlucky to discover that skeleton so anyway uh Pap and I and a lot of students uh gradually develop this picture that the mind is made of lots of processes or agents or resources or whatever you want to call them and it's anybody's guess what they are uh if you look at the anatomy of the brain uh you know that people label region so it's very clear that this occipital loob back here is largely concerned with vision but uh and I forget where the one for hearing is uh if you destroy the uh part of the brain for hearing in some animals you get a little bit of uh increased function in some part of the visual system that seems to enable the animal to hear a little bit and make some reactions and there's a whole lot of uh hype I think you have to call it about the flexibility of the nervous system that is if uh certain brain areas get destroyed other parts take over they almost never are as good and mostly they many functions never get taken over at all but are replaced by ones that superficially seem similar and so uh there's a whole lot of I guess wishful thinking that the brain is immensely resourceful and error correcting and repairing I think uh there was some idea that that was a general phenomenon but if you do some arithmetic you get an interesting result suppose that each function in the brain uh occurred in 10 different places at random then if you removed half the brain how many functions would you lose well almost no arithmetic tells you you would lose about one part in the thousand and so uh in fact you would never be able to detect it U so this idea that the brain has enormous redundancy will now change that number to five suppose each function is somewhat uh supported in five different parts of the brain then if you take off half the brain then what am I saying one part in 32 chance of uh of losing some significant function so probably uh lots of things that we do are supported in several parts of the brain apparently The Language Center is pretty unique uh and some others but uh be careful about the conclusions you read from optimistic neuroscientists anyway Pap and I worked on this idea of uh how could these large numbers of different processes been been organ be organized and we made various theories about it and uh then around I guess the late 1970s we stopped working together and paper developed his revolutionary ideas about education which um certainly have had a lot of influence although they didn't sweep the world and the way we had hoped and I kept working on the Society of Mind Theory and uh we didn't work together so much but we still did plenty of criticizing and supporting of each other anyway my theory ended up with this idea that uh it's sort of based on Freud I don't know if I kept a picture of his here Freud concluded that the mind was an inter interesting Arena sort of and he had the Mind divided into three parts there's at one end of the Mind the which were inherited from most other animals are is called the ID which is a bunch of instinctive mainly built-in behavioral mechanisms and um second part of the mind is what he called the super ego which is a collection of critics so in Freud's first image the brain is in two parts one is a set of instinctive built-in behaviors and the other is a set of critics which uh actually are associated with a culture or well with culture and a tradition and you learn from other people things that are good to do and things that are bad to to do and uh that's called the super ego this is your set of values and standards and tests for suitable Behavior and the middle is this strange object called the ego which is not what uh people think it is at least Freud's word the ego is a kind of big neutral Battleground where the uh instinctive behaviors I I keep wanting to oh you can see that Arrow if I take my finger off it um and then gradually as I kept trying to figure out where how problems are solved and what kind of processes might be involved I got this picture which has six layers and and various people come around and say I don't think you need to distinguish between layers four and five or uh why don't you just lump all the three top layers into one and I sort of laugh quietly and say these people are trying to find a physics like unified minimal theory of psychology and they're probably right in one sense but if they do that they'll get stuck because if you get a new idea there'll be no place to put it so if you have something that's very mysterious don't imitate the physicists because if you make a theory that's exactly right and just accounts for the data and there's nothing extra and nothing loose then when you notice a new phenomenon like dark matter uh then the physicists don't know what to do should they regard dark matter as some obscure feature of SpaceTime or does it have something to do with this universe being near another one that you can't otherwise communicate with and uh it's all very puzzling but there are lots of things that don't fit into Newton's Laws these days and uh I'm not suggesting a six layer Theory of physics but it might be worth a try okay so what am I I made up some nice slides but I think I'll stop so who has some questions and what would you like to see in a theory of psychology what's what do you want to be explained a lot of people uh are convinced that there are some really serious problems and Mysteries like what is consciousness and if you look at chapter 4 um my feeling is consciousness is an eological accident that people got a word which is a suitcase for all of the things they don't understand about the mind and more but once you've got a word and it goes in the culture consider the word Consciousness for a minute from a legal point of view suppose that you happen to be walking along and you're carrying something where is that pointer and it happens to stick somebody's eye out then it's very important when they sue you to establish whether you meant to do it or whether it was an accident did you consciously plan to uh I can't think of the English word for putting somebody's eye out there's beheading and all sorts douge is a good word um yeah so anyway uh it's very important for social reasons to have a word for whether uh an action was deliberately violating the rules as opposed to accidentally violating the rules like uh if you tripped on the stairs and landed on somebody and broke their neck that's not a crime unless you were so clever as to make it appear that it was an accident probably anyway you see what I mean so we need a our whole system of fairness and ethics and social responsibility is based on the distinction between whether an action was deliberate or not and so uh did he do that consciously is a word for that and somehow the idea of conscious became elevated well that's a very superficial you can probably think of 10 other reasons why a word like that yes Mar sometimes in your writing I uh you know seem it seems to me that that there's both sides of it some you argue with both sides of it like that it seems like in that kind of I I can well imagine there's that kind of representation you have a representation of self a representation of your mind but then you say there's no self or no consciousness why can't you think about Consciousness is just a process that is reasoning about your own mind that I I understand you don't want to talk about Soul that's a religious notion but no but sometimes when you say conscious that is do you remember doing it which is yeah but don't I load up my process that uh asks me what I think about my own mind I retrieve that I say yeah I do remember it well you're right actually I went to a lot of trouble to find uh 25 or 30 different uses for the word Consciousness and probably if I or if one of us worked harder we could take those and condense them into five or six much better ones that account for more stuff than the 30 well I think you got five right there ideas goals memory thoughts and feelings right that might be just right for something who knows yes well uh I think that's a great criticism of uh one reason why people don't like these theories quite so much because I propose too many things uh I really should reprint that uh that criticism from restac or handed out because it's this neurologist who says why is he telling us all these things about K lines and um representations and so forth the answer is he's from a community that that doesn't have enough variety yet and uh I'd be the first to admit that uh I try to go overboard and think of five more things than are in the literature but that emotion thing is nice remember that that was a serious challenge because when I made that list I do have a laser pointer somewhere in my jacket probably um how many how many words for at least noticeably distinct ways of thinking or reasoning or figuring out or solving problems can you think of you know maybe there are 20 or 30 I just I just realized this afternoon that I never looked I don't remember where I got this list yeah um can I ask about your conception of Free Will and I readings that you don't have a strong sense of will so what is that I think it's the same as the one for Consciousness namely it's a legal concept the idea of Free Will is completely obscene isn't it what could it possibly mean if you did something for no reason so it's a thoroughly empty idea isn't it or what do you mean by it do you mean there's nobody ordering you around so you you're free to do whatever you want but of course you're not you can only do what your computer computes you to do so in a sense like in the same way that you draw like the in the you show the fish diagram right the fish is is a the fish's actions are a product of the environment and of its current state and it's essentially a machine well it's some kind of machine yes sure and so would you argue that we are all that are just running out in the world sure uh I've never heard of any even interesting alternative in other words people who insist on free will appear to me to be like people who believe that there must be a God who created the world what's the next step who created the god they don't take that step so if your will is free okay then who's controlling it there's nothing there but but legally it's great because if somebody uh stole some many of their own free will but suppose you were a peculiar kind of epileptic and every now and then when you go [Laughter] by your hand goes out and steals things then they what do they do they they put you on parole no this is very strange but if you look at religions you see that people make money on them 133% of the world's product goes to people who make their living on Concepts like Free Will and Consciousness so it's a big money thing it's not it's not just an accident it's it's an industry so both of those are conct of society imagine a society without the concept of Consciousness in but those AR requirements for being I don't think you could you'd have to make up something to keep people in check and under control and uh to train them it's like the rat being the rat needs somebody to press the reward or punish button and we have it built into our a culture works because you build into people's head the machinery for suppressing doubt and it's very clever but but it you should think of it as an industry rather than a inexplicable phenomenon how much money goes into yes how many ways of thinking can you think of say it again how many ways of thinking can you think of right now that's my challenge in fact uh there's probably a big list in some chapter or other but but there's nothing like this what's the trick three if I go like that actually dragon has a thing so you can tell it make things bigger how many of you used the new Dragon program speech thing I can't believe how good it is oh I was talking to Henry Lieberman about it earlier yes so I'm the impression you would say that the side mind both humans and animals uh is it just that we have higher organization them so like where would you draw that line like do animals not have a notion of the self the way you describe in the like longm that's a great question and uh it'd be interesting to think about ways to investigate it people are researchers are often in fact raising that question of do animals have a representation ofs and there's a famous experiment but I can't remember what its current status is where you put a red dot on a chimpanze chimpanzee's head and when the chimp passes a mirror and sees that the chimp might go like that whereas uh I don't think a dog when it passes a mirror would uh rub its forehead to see if it has a red spot I had a cat who walked past mirrors because we have some wo full-size mirrors around the house and the cat walks by and there's this other cat in the mirror and she pays no attention to it whatever so of course I don't know what happened the first three times she walked by that mirror because if you see another cat do going by you'd think it would anyway it's it's a good question and people ask that and uh there's some evidence that elephants have a model ofs and maybe dolphins and I don't know where have any of you heard any stories of other animals that can recognize for example when they've been painted which elepants yes I think elephants there there's a famous child psychology where where if you're less than a couple months old you actually fail this test oh so it's actually something that that comes and is a sign of your child progressing so it might not be intrinsic to humanity it might be so I'm sure CH this is a off the topic but I once had a great email correspondence with some woman who was getting a PhD in France about when B how babies recognize their mothers and she concluded with she did experiments after having other people walk into the room with a mask of the mother or a different haird and so forth and for the first two months or three months I think um it turned out that the baby recognizes the mother by the haird which had not been known and then uh I think after 3 months it's recognizing the mother by a face and at that point uh she's doing experiments where you get another woman wearing the face a copy of the mother's face and so now there's two mothers and the baby is absolutely delighted and then as I I I can't remember but then I think at four or five months when two copies of the the mother comes in the baby gets really panicked so uh I lost track of her if if you have a baby let me know uh she got her PhD for this and um I haven't heard anything since yes here's a sort of half formulated thought that I just thought of um regarding ways of thinking and ways of feeling um so it just occurred to me that it seems like um if you go back to the list of feelings it seems like uh when we talk about feeling we're talking about a state that the brain is in so it might be a complicated State that's like some combination of a lot of different parameters but it's a state that you can stay in for like an arbitrary amount of time but I think thinking is um something that's more sequential as in like when you're thinking you're necessarily changing the state of your brain all the time is you're moving bids around so I don't know I just thought of no I think that's that's uh right that thing when we talk about intelligent Behavior you're absolutely right what you've got is a process that's criticizing itself and seeing when you got stuck and finding things to do and I suppose uh in each emotional state you're certainly also thinking uh so that's going on but but maybe it's more restricted like uh if you're confronting somebody and there's a sort of conflict uh then almost all your thoughts are constrained to to that subject and it's it's not as resourceful but I'm just I'm just improvising or perhaps like for each emotional state when we say that we're in an emotional state the thing that we talking about are like the certain switches that get flipped in a certain state or something is above a certain threshold and you can have thoughts that are about anything in that state but perhaps the state itself influences the type of thoughts that you're likely to have okay I think what I'm talking about in this context is sort of extreme forms where uh the person changes into another machine and it's like an angry person won't listen to reason so uh or it's very hard to deflect them and so it's kind of rigid thing so U but humans are generally are rarely in such extreme states where nothing gets through the whole point of that was that uh and I just realized that I maybe was just too lazy that we have this huge vocabulary of of nuances of emotional activity and people think these are also people think that these are hard to explain and mysterious and non-physical and blah blah blah but uh why do we wait till next week and see if somebody comes up with or see if we can come up with a set of 30 or 40 words about uh intellectual States curiosity of of of I just don't know how many there are and I I haven't thought of any in the last few minutes yes has anybody thought of a couple of the concept what's the word flow like CHS in the high like when you're really engaged in some activity that you're doing and like you're really in the in the zone uhhuh yes there's this there's a state of keeping other things out so you can focus not not being interruptible yes so in the book the side of mind you talk about agents and how like they divide between themselves but I don't I don't see anywhere about like how Evolution like modified a lot agents like I believe that Evolution modify the way we think right now I don't know if you saw there's a paper like uh wrote like two years ago it's called the region of behavior and this guy tried to explain like how we make some decisions evolutionary like in a point of view of like our speech we are uh maximizing the probability of we reproducing our our us but individually we are not kind of increasing the efficien and like I think somehow like these agents would be like the decisions of Agents or resources would be like very determined by Evolution since we have like a very long time to like of evolution of the human being and somehow we have like hard wire to make some decisions so uh for example like he gives the example of like guy well for for lizards that's certainly true but how why do you humans keep changing their environment and so like for example yeah so he gives this example of like tossing a coin the guy says that like this coin is unfair and like there's seven he doesn't say the probabilities but there is 75% of getting heads and 25% of getting uh fails and like we like the subjects they choose randomly 25% of the time uh the tailes even though they like they can take a uh account of the number of the times that you put heads and even though uh if you choose always heads you would get more money or whatever you would uh make it the right decision like that's called probability m and it's not a good strategy yeah but humans do that like no well they they do it if the psychologist rigs the experiment very carefully it turns out that the best thing to what do you think is the optimal strategy no the optimal strategy turns out it's the square roots of the probabilities normalized to add up to one and I I'll I'll give you a proof next time this theorem is due to Ray Solomon off who invented inductive probability Theory but but evolution so Evolution if it if evolution did probability matching then it would be wrong and I bet you'll find out that those experiments are wrong you have to see how did he rig the experiment so that people if it's probability 25 % they guess that 25% they me with fish I don't know there was a theory about why you would expect it but it's it's a good question I don't think people use probabilities though so yeah even if an experiment shows some I would look for a flaw in the design of the experiment yes so you talked about emotion my understand is that like um for for someone who has a specific personality they might have predisposition to feel a certain emotion like anger depression or whatnot um my my question is that if you if you had any like Insight or theories on to what extent personality is uh affected by like events or influences that happen to us over the course of a lifetime and to what extent is it impacted by sort of you know chemical or biological makeup of well you're asking uh what things do what do people learn we don't care if it's chemical or see if it's chemical it's still physical because I mean um I've read a little bit of that like uh sort of treatment for depression and the argument is that a lot of a lot reasons for depression is because of some sort of chemal biological we are brain constructed well there's lots of complicated things about the brain one feature of the brain that I don't know if everybody you know that there are inhibitory and excitatory synapsis when one neuron connects to another um the impulse that goes along the axon to the Target neuron May reduce the probability or the strength of its firing or increase it so that's called inhibiting or there's no or exciting it's not quite the right word is it um now generally in the nervous system as a rule but not always if you follow a chain of activity it goes inhibiting exciting inhibiting exciting if you had too many excitatory things and there was a loop then it would explode and they it would wear itself out in jig time so there is this General feature of the anatomy that you alternate so when somebody talks about a drug having an inhibitory effect that's sort of weird because it's inhibiting half the neurons and uh therefore ex raise lowering the thresholds of the ones they're connected to and so on so I think the best thing is until you have a diagram of the functional relations between different brain centers uh it might be best not to try to make generalization about how the chemistry works it's it's easy but you know people think of adrenaline as a stimulant but in the nervous epinephrine but in the nervous system it's uh locally it may be inhibiting things that are inhibiting something else and so it appears to be exciting yes so I have a problem understanding the difference between thoughts and emotions I know it might be a simple but uh since the only thing that I can separate in my mind is that thoughts the so let's call it a time constant I can change it kind of rapidly emotions time constant is small and I can control it again much less uh but since there is no um at least in this class there is no free will uh how can I how can I make a decision on dividing line between these two entities oh I think uh it's a waste of time as far as I can see emotional mechanisms are generally lower level simpler ones than the ones that involve several layers of the more layers of the brain so it's just a relative thing it's it's not that some states are emotional you're always having some high level thoughts and low-level thoughts and the distinction I just don't know don't know why the distinction has occupied so much attention I think it's because and that goes back to the having more words for or asking how many words do we have for ways to think it seems to me that in popular culture there are very few words for ways to think and lots of words for emotions and so they're very prominent maybe you have to be smarter to distinguish between ways to think and people generally are dumb not because they're inherently dumb but they come from cultures which bully you if you you know what happens if you're in third grade and you're smart you get it beaten out of you and you learn not to show it question why science happen earlier why don't we have more ways to describe different ways to think and that sort of we just not reflected as much on different thinking States or different approaches I wonder if the Greeks had more but we I think I think they did I think they had also More Concept of ideas in different uh different states different potencies H who has a theory of that what's your theory of the Middle Ages how could things get dark for so so dark for so long well I do have a um and are we about to have one I think that's much to do with the channels in which one can communicate ideas to other people whether they exist or not um whether the are open or closed um um the Middle Ages were characterized by scientific discoveries being kept as family secrets ah cardan knew how to take cube roots and he wouldn't tell anyone well the classic stupid example is baby Tom which for 300 years made a single Italian family very rich you know using tongues to extract a baby and child bir increased success rate in difficult births by about 10% they say wow and that was enough to build a family forun until some servant part stole the beam and that was the end of that oh wow so who has a theory of the Middle Ages is there a standard theory yes well the concept of the Middle Ages as the Dark Ages is something that emerged mostly in the Renaissance when people in the 14 well 15th 16th century tried to present themselves as going back to the classical age of scholarship of ancient grece and Rome and as being better than their predecessor for the last few years this mostly happened because of discovery of manuscripts that were translated from the ancient Greek in certain cases Latin by by Muslim uh who at the time were sort of receding from Europe but uh so the entire con of the Middle Ages might be a fabrication of the Renaissance I mean there were some significant discoveries at the time that's a nice idea in other words when was St Patrick St Patrick I'm told that he he uh popularized a lot of technical manuscripts wrote them to back into Europe which is he has two achievements one was in scientific culture back and the other was getting snakes out of England or something Ireland I don't know which he was sainted for don't you have to do three Miracles or is it what's what's yeah yeah my theory is that the Middle Ages ended around 2100 because they'll say after the Middle Ages end they'll say you know those guys back in the 21st century you know they had no idea how things thinking work they couldn't even think of a few ways to think they had poverty they had Wars you know those guys were barely out of their MO CL right I just read a history of AI I forget who wrote it but it had this section saying mentioned the newal Simon there was a thing called General Problem Solver which I mentioned a couple of times in the book and it's the idea that the way to solve a problem is to find it's a symbolic Servo find the difference between what you have and what you want and look in your memory for something that can reduce that difference keep doing that and of course it's it's important to pay attention to the more important differences first and so forth and I'll send you this article this article is saying that that they made a terrible mistake and this was a trivial Theory and that's why nobody uses it anymore and it was interesting how many AI people fell for that idea in the 1960s my complaint has been that uh if you look in a modern textbook on you must have some in your first volume Pat didn't you have some GPS things if you want to keep up with AI you should read Patrick's textbook even though he's people are starting to use this new one which doesn't have any AI in it by who is it by Russell and Russell and norig it's probably pretty good technically but I Leaf through it and it didn't have any never mind it's probably better than I think because I'm jealous yes kind of a different topic in Society of Mind you're talking about the amn infancy when you forget what you learned things that were become common sense and you can't even remember how it was you learn that so it's just of that process when you try and say teach something to somebody that you turn that you're putting Awareness on and bring bringing back up different levels not question just wondering about is that another way learn does that answer that's sure an interesting question when I first learned about programming I was inter I had the idea that maybe babies think in machine language and then after a while they start to think in Fortran and then finally when they're a little older they think an Alo or something and so uh but when they switch from machine language to Fortran then they can't remember their earlier thoughts and there's almost no evidence of genu people finding genuine Recollections from uh twoyear olds at later ages now almost everybody thinks they remember something but there's the problem that you might have rehearsed it and translated it into the Fortran and the algol and the lisp and the logo go you know whatever whatever it is I had one of my greatest influences was uh uh a great mathematician named Andrew gleon at Harvard who I met practically the first day I got to Harvard and he would always talk about things I didn't understand and I would go home and look them up and try to anyway uh one day we were talking about number forms and number forms are a psychological phenomenon which about 30% of people have and it was first described by Francis gton and the pro the phenomenon is if I ask you close CL your eyes and tell me where is the number three how many of you in have a place for the number three well that's a few and so typically if you imagine the visual field that's a windshield I guess so there are these numbers and they're they're nowhere in particular except that it's usually like that for a older child so here are these numbers and what's more in some people they're colored so I was talking to Andrew I had read this goon paper which is 1890 or 1885 or something and so I was asking asking people if they had number forms and he said oh yes he has one and uh he sketched it for me and he said and they're colored too oh and his went way up and the prime numbers were bright I what am I doing maybe the composite numbers were something was bright and they were colored so I wrote this down and I over the next couple of years I would look in antique stores for old children's blocks and I found a set of blocks that matched that and Andrew Gleason said that he knows when he acquired this thing and it was about 4 years old and he had a window which in his house and there was a hill and he could just read he could just see over the sill and he imagined these numbers on the side of that Hill blah blah blah anyway uh people who don't have a number form don't know what I'm talking about but um and I don't know if the 30% is still true but um it's an interesting phenomenon and in most cases you of Early Childhood well that's you can't find out because CH children do remember details of a house they lived in but you don't know if they've copied it so what was the original question how much can you remember from infancy elran haard thought you could go back to before you were born and uh you could remember people talking about you when you're still in the [Music] womb so anyway John Campbell said you should look into this and uh a few of us made an expedition we went down to Elizabeth New Jersey to visit the the just starting up diic Center and I met this elron hubard who had green eyes and was quite hypnotic looking and the end of the story is he had been writing about how if you took this treatment of Dione XS then you could memorize an entire newspaper in five minutes and do all sorts of miraculous things like that once your mind has been cleared of ab ations and obstacles and it became a big industry and turned into later Scientology I'm sure you've all heard about that so we asked hubard to look at a newspaper and tell us what was in it and he explained he was so busy training the other people to be cleared that he hadn't had time to go through the procedure himself and I never saw him again yes what are your thoughts on and the fact that we're just replicas our thoughts repli of memes you mean Dawkins I didn't quite get the whole question the idea that uh oh so me so for example the way we talk and the way we all talk how we very much remic the way our parents talk for people around us um and possibly the way we think as well so how does that relate to how our mind develops are we actually creative original characters well of course it's both because you learn things from your culture and then uh you might just mainly repeat things or you might get the Knack of making new ideas I'm trying to remember what Dawkins main ideas are uh he invented the word meme to say that the ideas that people have my might be considered to be somewhat similar to the genes in our heredity and that societies are systems in which these memes which are uh conceptual units of meaning or knowledge uh propagate around and self- reproduce and mutate and spread and I don't know what to say about it except that it's obviously true that uh every now and then someone gets a new idea and tells people and for one reason or another they either forget it or or tell someone else and after a while it spreads and some of them fill up the whole culture and some just die out and whatever else Dawkin says he's very smart guy but almost everything then uh in my mind is that he he's explaining that religions are mostly made of these memes and they're very bad and cost the world a great deal in uh progress and productivity in other words he's a militant atheist and it's there about five bests sellers in that business today but I don't know what else to say about memes it's an obviously uh General correct idea and but the great thing about genes is we know the four Amino a or four nucleic acids they're made of and how they're roped together and all that and I don't think Dawkins Theory uh develops anywhere nearly as elegantly as modern genetics so it would be nice if it could but a really good theory of good ideas would be nice to have what would it what would it look like someday we'll have an AI that just punches them out I think I think it would look like a really good language uh oh right Robert heinlin has some stories in which the super intelligent people have have a language that's so dense that in five syllables they can explain something that would take you a half hour I think I forget what it's loglan made anyway if you need a good idea read Robert hline sure um I where you mentioned this you talk about how geniuses have just come up come up with um better ways to um think better like better way to better ways to learn yes how to learning but why do you think um they never and why hasn't they like meod learning about that's a great question and do think have like you know a concept of an idea that like improve better learning that there's a couple of phenomena that like how come there were so many Geniuses in Athens and that's and then uh some of the best mathematicians came from some high school in some little country next to the Baltic Bulgaria I can't Romania ran yes there was some high school in Romania that not only produced van noyon but about five or six other worldclass mathem itions I don't remember the details so that's a nice question how come uh if there are these great memes how come there aren't more big pockets of them but there a lot of cultures which were very inventive in in other than intellectual Fields how come Paris got all those artists in the how many of you saw the Woody Allen movie what's it called Paris at midnight it's so funny right that's a good question for each of us what's your very best IDE idea and stop fussing with the other ones and get that one out there must be some people who are very quiet and only speak once in a long time we should watch them carefully yes I guess along those lines do you ever feel restricted by language because every IDE wait it's the sound I can't hear from is that strong enough to lift [Laughter] you I'm sure someone's done it sorry about that um oh do you ever feel restricted by language and that you must represent your theory of mind or any idea language uh no I don't but I I once was jealous when paper explained that he got he got some idea and he explained he gets ideas like that when he thinks in French and and uh you draw pictures too what you draw pictures too oh yeah so it's not just language that's a language Graphics so we ought to have devices within the next few years that um draw pictures when you think or it's so funny you know uh we had cyborgs what were they called I mentioned them last time we had Steve man and who's the other one yeah so there's these two guys around the media lab wearing various things on their head and and they're always typing and you ask them a question and they've searched Google and uh when was that 1990 but it's all gone nobody walks around with direct connection to the web St still does yes anyway I certainly expected it to turn up and something wrong so anyway you should be able to buy one one of these days and uh what's her name who was that nice woman who had the EEG thing you remember at Ted forgot her name anyway she had a sort of helmet which had about 20 electrodes and she induced me to put it on I was on a stage with about a thousand people there which was rather funny and there's a little spot on a CRT and I get to think about it moving one way or the other and rewarding it when it did the right thing for only about half a minute and then I could steer it around so here was a nice uh primitive Gadget where you could sort of almost draw just by thinking this spot and then she started a company and hasn't sent me one because maybe it was just beginner's luck beginner's luck or something what her name yes Lee yeah uh did you find the company but you think that there be lots of people wearing stuff and she's doing it right with your keyboard right why do we take a five minute break I don't know huh well I hate to interrupt because I see 10 different productive discussions yes I I see a I saw quite a few apparently productive discussions maybe that's what the class needs but uh anybody come to to a conclusion yeah see if you can knock the wall down and yeah it's not a conclusion but it's a question it's a thought experiment so if you had a black box that could uh sort of replace part of your brain and let's say you replace like 5% at a time at what point if you assume that there's this self entity at what point would you lose yourself you if you change the question is are you how how much do you have in common with uh with the youu of yesterday as compared to uh when you graduated grade school so this question of the idea of identity is very very fuzzy yeah so that question sort of of these peculiar cases of transplant Pati suddenly having preferences like the person they got the trans problem like heart transs and stuff Som they they start to like same Foods or use the same words things like that the same spouses I read a a science fiction novel by Robert Sawyer any of you know of him Canadian writer and uh it has to do with somebody who has a fatal disease so he's going to die soon but the technolog is around where you can make a duplicate of him and uh so he has a duplicate made and he is sent to the moon for some reasons I can't remember which is a kind of a nursing home for I think people who are enfeebled and do much better with 17th gravity and so there's some reason why anyway the the living the original copy is sent to the moon and the substitute takes over but then uh our hero is miraculously cured by uh eating the right stem cells or what I don't remember so now the so he wants to come back and the question is who gets the car so I can't remember the title of the novel except that it has alien in it alienable rights is not something like that but I wrote an article called alienable rights but anyway that's so are you the same as you were five minutes ago or five years ago or whatever and as far as I'm concerned the answer is who cares uh it's a sort of silly question because no two things are exactly the same ever anyway and but again a lot of these questions which look philosophical are legal so the joke of uh that novel is that who owns the car is what matters to decide who is the real uh original and who's the copy Frederick pole wrote a similar story much longer ago where people are copied and the copy is sent on a oneway trip to some planet to fix a broken reactor and they always die and you get a million dollars for providing this copy but one of the copies survived so it's the same plot I can't remember that if you're looking for a good idea if you go to 1950s science fiction look for AE van vooder Frederick Poler all those wonderful writers that was before it was necessary to describe really good characters and science fiction got better and better for the literary critics and generally worse and worse for the science fiction fans do we really have any more questions yes so we think in age where uh sharing information between many people um uh short short amount of time is quite easy internet and anything like this do you think this will change uh this will bring up any more ideas do you think that this will hum because before before this time people had like problems with sharing the information and also now it's much easier to have like a large group of people working on one thing do you think that this will change the way that we think and do you think that this will it's a tough one uh bad things can happen and good things can happen and but that's funny because that reminds me again of Science Fiction because in science fiction many many years ago some writers got the idea that there would be something like an internet and some people realized there would be flash crowds and now there are flash crowds and I remember even as a kid talking to people who said why not wire up the voting machines so that they're always there and so if the government somebody in the government wants to know uh should we do this or that uh should we bomb China or not uh you could get a 100 million people to run up to the keyboard and say yes or no and presumably when the what do they call those that great crowd of Jefferson and Franklin and founding fathers the founding fathers did a lot of things to prevent that and the one that they focused on which was one of the most effective was called The Electoral College and the United States is different from other places because we don't elect congressmen or presidents we elect smart people from the community who then get together and decide who should be president and of course now if anybody now they belong to parties and if any of them voted for the other Party's candidate they would be held to pay but it was a great idea because the founding fathers realized that if you had instant feedback which is what Hitler got uh then you could say something really exciting and everybody would press the yes button and then you kill all the Jews and then uh the next speech you kill all the black people and all the yellow people and all the people whose last name doesn't begin with them and so uh what you don't want is instant feedback now the new social networks are getting us close to that and the question is is it time to have is it time to stop that is it getting dangerous I don't know but there must be a lot of people who are recognizing that this thing is creeping up on us and you might be able to get 50 million people to do something reckless in a few minutes if you if you don't uh put some limits I don't think we could get the the Electoral College back because you'd have to get a majority to what does it take to fix the Constitution 2/3 we'll never see 2/3 again it's the end of America well we have three minutes yes if you design a direction for the field of psychology obviously went more than just set of debugging tools what would you say they should be doing they should read Patrick Winston's thesis the psychologists now have disappeared into the tarpit of statistics and they don't have the idea that knowledge needs complicated representations and I don't care whether you assign probabilities to them or put them in the order in which you thought of them or you know or do what doug lenett did in his thesis of swapping things when one worked better than another but but uh I forget the question but I think we we've got to get better ideas about representation of knowledge and I don't know where they're going to come from now that uh the whole uh AI Community is drifting into these ways of avoiding representations I haven't read the norvig Russell book so does he can anybody summarize what it says about knowledge representation who's read it was it there's a chapter on logic logic that's so funny first order logic is what new and Simon thought of in 1956 before they thought of the so-called GPS thing logic can't make analogies it's a very bad thing to get stuck with zero or one maybe one of our papers should be on what should AI do next year so really what my main concern it has been for quite a few years uh is to make some theory of how uh what makes people able to solve so many kinds of problems I guess if you ran through the spectrum of all the animals you'd find lots of problems that some animals can solve and people can't like uh how many of you could build a beaver dam and or a termite Nest so there are all sorts of things that Evolution manages to produce but maybe the most impressive one is the what the human infant can do um just by hanging around for 10 or 20 or 30 years and uh watching what other humans can do so we can solve all sorts of problems and uh my quarrel with the rest of the most of the artificial intelligence Community has been that uh the great success of Science in the last 500 years really has been in physics and it's been uh rewarded by finding little sets of rules like Newton's three laws and Maxwell's four laws and uh Einstein's one law or two that explained a huge range of everyday phenomena of course in the 1920s and 30s that apple cart got upset because actually uh Einstein himself who had discovered the first Quantum phenomena namely the uh quantisation of photons uh had produced various uh scientific laboratory observations that were inexplicable uh in terms of either Maxwell or Newton or Einstein's earlier formulations so uh my picture of the history is that in the 19th century and little bit earlier going back to lock and Spinosa and Hume and a few of those philosophers even Emanuel K they had some pretty good psychological ideas and as I mentioned the other day uh I suspect that Aristotle uh was more like a modern cognitive psychologist and had even better ideas but uh we've probably lost a lot of them because there are no tape recorders who knows what Aristotle and Plato Plato said that their students didn't write down because it sounded silly uh so um the idea that uh we developed around here mostly uh SE more paper and uh a lot of students Pat Winston was one of the great stars of that period was the idea that uh to get anything like human uh intellectual abilities you're going to have to have all sorts of highlevel representations so one has to say the old conditioned reflex of stimulus versus stimulus producing a response uh isn't good enough you the stimulus has to be represented by some kind of semantic structure somewhere in the brain or mind and so far as I know it's only in the theories of not even modern artificial intelligence but the the AI of the 60s and 70s and 80s that people people thought about what the what could be the internal representation of the kinds of things that that we think about um and uh even more important if one of those representations you see something or you remember some incident and your brain represents it in some way and if that way doesn't work you take a breath and you uh sort of stumble around and find another way to represented maybe when the original event first happened uh you represented it in three or four ways and so we're beginning to see uh did anybody hear fi's talk uh the Watson guy was up here a couple of days ago I missed it but um they haven't made a technical publication as far as I know of how this Watson program works but it sounds like it's something of a interesting Society of Mind likee structure and it' be nice if they would has anybody read any long paper on it there have been a lot of press reports have you seen anything Pat Uh so anyway they seem to have uh done some sorts of Common Sense reasoning as I said the other day I doubt that Watson could understand why you can pull something with a string but you can't push and uh actually I don't know if any existing program can understand that yet I saw some amazing demonstrations Yesterday by um or Monday by uh wolf Steve Wolfram of his wol Alpha which doesn't do much Common Sense reasoning but it has what it does do is uh if you put in a sentence it finds five or 10 different representations anything he can find that's sort of mathematical and so when you ask it a question it gives you 10 answers and it's much better than previous systems because it doesn't uh well Google gives you a uh a quarter million answers but that's too many and anyway I'm just going to talk a little bit more and just everybody should be trying to think of a question that the rest of the class might answer so there are lots of different kinds of problems that people can solve going back to the first one like which moving object out there is my mother or and which might be a Potential Threat so there are lot of kinds of problems that we solve and I've never seen any discussion in psychology books of what are the activities principal activities of Common Sense thinking somehow uh they don't have or people don't before computers there really wasn't any way to think about high level thinking because there weren't any technical technically usable ways to describe complicated processes the idea of a conditional uh expression was was barely on the threshold of of psychology so what kind of problems do we have and if you take some particular problem like uh I find these days I can't get the top off bottles so how do I solve that and uh there are lots of answers uh one is you look for somebody who looks really strong or you reach into your pocket and you probably have one of [Music] these and so on there must be some way to put it on the floor and step on it and kick it with the other foot but uh so there are lots of uh problems that we're facing every day and uh if you look in traditional cognitive psychology or well what's the worst Theory the worst and the best theory was got popular in in the 1980s it was called rule-based systems and you just have a big Library which says if you have a soda bottle and you can't get the cap off then do this or that or the other and so some people uh decided Well that's really all you need Rod Brooks in the 1980s uh sort of said we don't need those fancy theories that people like Minsky and paper and Winston are working on why not just say for each situation in the outer world uh have a rule that says how to deal with that situation let's make a hierarchy of them and he described a system that sort of looked like the priority interrupt system in a computer and uh and he won all sorts of prizes for this really bad idea that's spread around the world and but it solved a lot of problems uh there are things about priority interrup that aren't obvious like suppose you have uh in the first computers there was some problem because uh what should you do if there's several signals coming into the computer and you want to respond to them and some of the signals are very fast and very short then you might think well I should give the highest priority to the signal that's going to be there the the shortest time something like that the funny part is that when you made such a system the result was that if if you had a computer that was responding to some signal that's coming in at a uh I'm talking about the days when computers were only working at a few kilohertz few thousand oper ation a second God that's slow million times shorter than what you have in your pocket and uh if you give priority to the signals that have to be reacted to very fast then what happens if you type to those computers it would never see them because it's always I saw this happening once and finally somebody realized that uh you should give the highest priority to the uh to the inputs that come in most least frequently because there's always otherwise if there's something coming in very frequently you'll just always be responding to it any of you run into this um took me to while to figure out why uh anyway there are lots of kinds of problems and um the other day I was complaining that we didn't have enough ways to just we had hundreds of words for emotions and uh here's a couple of dozen they're in chapters seven and eight actually most of these so here's a bunch of uh words for describing ways to think but they're not very technical so you can talk about remorse and sorrow and blah blah blah hundreds and hundreds of words for for feelings and it's a lot of effort to find a dozen words for uh for intellectual for what should I call them problem solving processes so it's curious to me that the the great field called cognitive psychology has not focused in that direction anyway here's about 20 or 30 of them and you'll find them scattered through chapter 7 and 8 here's my favorite one and I don't know of any proper name for it but uh if you're trying to solve a problem and you're stuck and the example that comes to my mind is uh if I'm trying to remember someone's name I can tell when it's hopeless and the reason is that for somehow or other I know that there's a huge tree of choices that's one way to represent what's going on and you know I might know that I'm sure that that letter that name has a z in it so you search around and uh try everything you can but of course it doesn't have a z it's uh so your so the way to the way to solve that problem is to uh give up and then a couple of minutes later uh the name occurs to you and you have no idea how it happened and um so forth so anyway the long story is that uh p and I and lot lots of lots of uh really great students in the 60s and 70s uh spent a lot of time making little models of problem solvers that didn't work and uh we discover that uh you needed something else and we'd put that in uh other people would come say that's hopeless uh you're putting in more things than you need and and my conclusion is that uh wow it's the opposite of physics in physics you're always trying to find you don't want to what is it called aam's razor never have more structure than you need because because what uh well it'll waste your time but my feeling was I never have less than you'll need but you don't know how many you'll need so what I did I had four four of these and then I forced myself to put in two more and people ask what's the difference between self- models and self-conscious processes and I don't care uh or what's the difference between self-conscious and reflective I don't care and the reason is that wow it's nice to have a box that isn't full yet so uh if you find something that your previous Theory going back to Brooks uh he was so successful getting simple robots to work that he concluded that the things didn't need any internal representations at all and for some mysterious reason the artificial intelligence Society gave him their annual big prize for for this very wrong idea and it caused AI research to sort of half collapse in places like Japan who said oh rule-based systems is all we need anybody want to defend him the odd thing is if you talk to Brooks he's one of the best philosophers you'll ever meet and he says oh yes of course that's wrong uh but it helps people do research and get things done and as I think I mentioned the other day when the Three Mile Island thing happened there was no way to get into the reactor that was 1980 and 30 years later when the what's how do you pronounce it Fukushima uh accident happened there was no robot that could go in and open a door and uh I don't know who to blame for that maybe us but my picture of the history is that the places that did research on robotics there were quite a few places and for example Carnegie melon was very impressive in getting the Sony dogs to play soccer and they're still at it and I think I mentioned that Sony still has a stock of uh what's it called iOS say it again fibos fibo IO AO oh right I ibows right um but the trouble is they're always broken and uh we had a there was a robot here called Cog that Brooks made and it sometimes worked but usually it wasn't working and so only one student at a time could experiment with the robot what was that wonderful project of trying to make a walking machine for four years in uh there was a project to make a robot walk and there was only one of it so first only one student at a time can do research on it and most of the time it's something's broken and you're fixing it and so you end up that uh you sort of get five or 10 hours a week on your laboratory physical robot uh at the same time uh Ed fredkin had a student who tried to make a walking robot and it was a stick figure on the screen and uh forgot the student's name but anyway uh he simulated gravity and a few other things and in a couple of weeks he had a pretty good robot that could walk and go around turns and bank and uh if you put if you simulated an oily floor it could slip and fall which we considered the high point of the demo actually so there let me find so anyway I've sort of asked you to read my two books for this course but that those are not the only good uh texts about artificial intelligence and if you want to dig deeper uh it might be a good idea to go to the web and type in eron slowman as s l o m an and you'll get to his website which is something like that and uh slowman is a sort of philosopher who can program um there are a handful of them in the world and uh he has lots of interesting ideas that nobody's gotten to carry out and uh so I recommend who else is Pat do you ever recommend anyone else no what I'm trying to think I mean if you're looking for lot for philosophers uh Dan Dennett has a lot of ideas but but slowman is the only person I'd say is a sort of real professional philosopher who uh tries to program at least some of his ideas and uh he has successful students who have made larger systems work so um if you get tired of me and you ought to uh then go look at this guy and see who he recommends so okay any who has a good question to ask Mar you were talking about how we have a lot of words for emotions why do we only have one word for cause it's a mystery but uh I spent uh most of the couple of days making this list bigger but these aren't you know these are things that you do when you're thinking you make analogies uh if you have multiple goals you uh try to pick the most important one or uh in some cases uh if you have several goals maybe you should try to achieve the easiest one and there's a chance that it'll lead you into what to do about the harder ones but uh a lot of people think that uh mostly in England that logic is a good way to do reasoning and that's completely wrong uh because in logic first of all you can't do analogies at all except at a very high level it takes four or five nested quantifiers to say a is to be as see as to which of the following five or uh so I've never seen anyone do analogical thinking uh using formal logic first order or higher order predicate calculus um what's logic good for it's great after you've solved a problem because then you can formalize what you did and see if some of the things you did weren't necessary in other words after you've got the solution to a problem which you got by going through a big search you finally found a path from A to Z and now you can U see if the assumptions that you had to make to bridge all these various little gaps were all essential or not so um yes like what kind of example would you say that logic can do analogies like like water is for b as like containment or like why why well because you have to make a list of hypotheses and then let me see if I can find Evans the trouble is darn Evan's name is in a picture and word can't look inside its picture can power can PowerPoint find words in its illustrations why don't I use PowerPoint because I've discovered that PowerPoint can't read pictures made by other programs in the Microsoft Word Suite the drawing program in word is pretty good and then there's an operation in Word word which will make a PowerPoint out of what you drew and it it's 25 years since Microsoft hasn't fixed the fatal errors that it makes when you do that in other words I don't think that the PowerPoint and word people communicate and they both make a lot of money so that might be the might be the reason where was I well you can do anything in logic if you try hard enough but but uh a is to B as C is to X is a four-part relation and you'd need a whole pile of quantifiers and how would you know what to do next yes um Talk a bit about um the situation in which we are able to perform some sort of action like really fluently and really well but we cannot describe what we're doing um and the example I give is um say I'm like an expert African drummer from Africa and I can make these like really complicated rhythms but if you ask me what did you just do like I have no idea how to describe it and in that case do you think the person is capable of like or I guess do you think the person we can say that the person understands this even though they cannot explain it well but if you take an extreme form of that uh you can't explain why you use just any particular word for anything there's no reason it's remarkable how well people can do in everyday life to uh tell people how they got an idea but when you look at it it doesn't say how you would program a machine to do it so there's something very peculiar about the idea that goes back to this this idea that people have free will and so forth suppose I say uh look at this and say this has a constriction at this point why did I say constriction uh how do you get any how do you decide what word to use for something you have no idea uh so it's a very general question it's not clear that the different parts of the that the the frontal loes which might have something to do with making plans and analyzing certain kinds of situations have any access to what happens in the broker or what's the what's the speech production area broker and uh I'm trying to find the name of the other one it's connected by a cable that's about a quarter inch thick there's also ver Veri yeah we have no idea how those work as far as I've never seen any publication in Neuroscience that says here's a theory of what happens in W's area have any of you ever seen one what do those people think about but they'll tell you about I was reading something which said it's going to be very hard to understand these areas because each neuron is connected to a 100,000 Little Fibers well some of them are and I bet they don't do much except sort of set the bias for uh some large collection of other neurons but but we if you ask somebody how did you think of such a word they will tell you some story or anecdote but they won't be able to describe some sort of procedure which is say in terms of a language like lisp and say I ConEd this and that and I took the cter of this and the car of that and I uh put them in this register and then I swapped that with you don't see theories of how the mind works in Psychology today the only parts are they know a little bit about some aspects of vision because you can track the paths of images from the retina to the what call to the primary visual cortex and people have been able to figure out what what some of those cortical col cols do and uh if you go back to an animal like the Frog then researchers like bitsy and others have figured out how the equivalent of the cerebellum in the Frog uh they've got almost the whole Circuit of how when the Frog sees a fly it manages to turn its head that way and stick its tongue out and catch it but in the case of a human I've never seen any theory of how any person thinks of anything so there's artificial intelligence which has highlevel theories of semantic representations and there's NE Neuroscience which has good theories of loc some parts of locomotion and some parts of sensory systems and to this day there's nothing much in between uh so uh David here has decided to go from from one to the other and um former student of mine Bob Hearn has done a little bit on both and I bet there are 20 or 30 people around the country who have trying to bridge the gap between symbolic artificial intelligence and uh mappings of the nervous system but it's very rare and uh I don't know who you could ask to get support to work on a problem like that for five years yeah so presumably to build a humanik artificial intelligence we need to like we need to perfectly model our own intelligence which means that we're the system we ourselves are the system that we're trying to understand well it doesn't have to be exact I mean people are different and uh typical person has looks like they have 400 different brain centers doing slightly different things or very different things and we have these examples in many cases if you lose a lot of your brain uh you're very badly damaged and in other cases uh you recover and uh become just about as smart as you were probably a few cases where you got rid of something that was holding you back but it's hard to prove that so we don't need to we don't need a theory of how people work yet and the nice thing about AI is that we could get we could eventually get models which are pretty good at solving what people call everyday Common Sense problems and probably in many respects they're not the way the human mind works it doesn't matter but once you've got if I had a program which was pretty good at understanding why you can pull with a string but not push uh then there's a fair chance you could say well that seems to resemble what people do I'll do this a few psychological experiments and see what's wrong with that theory and how to change it so at some point uh there'll be people making AI systems comparing them to particular people and trying to make them fit the trouble is nowadays it takes a few months to if you get a really good new idea to program it um I think there's something wrong with programming languages and what we need is a we need a programming language which uh where the instructions describe goals and then sub goals and then finally you might say well let's represent this concept by a number or a semantic network of some sort but uh yes that idea having a program in language where you define goals is there a goal oriented language so there there's kind of one if you think about it if you squint hard enough um something like SQL um where you you tell the computer you know I want to find you know the the top 10 people in my database with this high value and then you don't worry about how the system goes that in a sense that's you're defining the goes you got a bit uhhuh what's it called sequel SQL at a datab oh right yes I guess uh database query languages are on the track but wolf from alpha seems to be better than I thought well he was running it and he Steve wlr was giving this demo at meeting we were at uh on Monday and he'd say well maybe I'll just say this and and it always worked so so maybe either the language is better than I thought or Wolfram is better better than I thought at something remarkable guy yes so I like this example of uh you only remember a name after you can remember a name after you've given up consciously trying to think about it do you think this is a matter of us being able to set up background processes and then either there's some delay like we give up there's some delay in the process where we don't have the ability to correctly terminate processes do you think this only works for memory or could it work for other things like could I start an arithmetic operation and then give up and then it'll come to me later well there's a lot of nice questions about things like that how many processes can you run at once in your brain and U I was having a sort of argument the other day about uh music and I was wondering if I see a big difference between Bach and uh the and uh the composers who do Counterpoint Counterpoint you usually have several versions of a very similar idea maybe there's one theme and you have it playing and then another voice comes in and it has that theme upside down or a variation of it or in some cases exactly the same and that it's called a Canon and um so the T to force in classical music is when you have two or three or four versions of the same thought going on at once in different times and my feeling was that in uh popular music or reg uh if you take it a typical band then there might be four people and they're doing different things at the same time uh usually not the same U musical Tunes but there's a rhythm and there's uh a Tony and there's various instruments doing different things but there you don't have several doing the same thing I might might be wrong and um somebody said well some some popular music has a lot of CounterPoint I'm just not familiar with it but I think that's if you're trying to solve a hard problem it's fairly easy to look at the problem in several different ways but what's hard is to look it in several almost the same ways that are slightly different because probably uh if you believe that the brain is made of Agents or resources or whatever uh you probably don't have duplicate copies of ones that do important things because that would take up too much real estate anyway I might be completely wrong about uh Jazz somebody um maybe they have just as complicated overlapping things uh as Bach and uh the contrapuntal composers did but yeah what is the ultimate goal of artificial intelligence is this some sort of application or is it more philosophical oh uh everyone has different goals or On's in your opinion uh I think we're going to need it because uh the disaster that we're working our way toward is that people are going to live longer and uh they'll become slightly less able and so you'll have uh billions of 200y old people who can barely get around and there won't be enough uh people to import from underdeveloped countries to or they won't be able to afford them so we're going to have to have machines that take care of us of course that's just a Trent because at some point then you'll download your brain into a machine and fix everything that's wrong so we'll need robots for a few hundred years or a few decades and uh then we'll be them and we won't need them anymore but it's an important problem what's going to happen in in the next hundred years you're going to have 20 billion 2 200 year olds and nobody to take care of them unless we get AI nobody seems particularly sad about that how long oh another anecdote I was once giving a lecture and talking about people living a long time and nobody in the audience seemed interested and I'd say well suppose you could live 400 years and most of the people I then I asked what what was the trouble and said wouldn't it be boring so then I tried it again in a couple of other lectures and uh if you ask a bunch of scientists how would you like to live 400 years everyone says yay and you ask them why and they say well I'm working on a problem that I might not have time to solve but if I if I had 400 years I bet I could get somewhere on it and the other people don't have any goal that's my cold-blooded view of the typical non-scientist there's nothing for them to do in the long run who who can think of what should people do what's your goal how many of you want to live 400 years wow must be scientists here try it on some crowd and let me know what happens are people really afraid yeah I think I think maybe the different factor is whether or not your 400 years is just going to be the repetition of 100 years experience or if it'll start to like take off you'll start to learn better progress right I've seen 30 issues of the big bang and I don't look forward to the next one anymore because they're all this they're getting to be all the same although it's the only thing on TV that has scientist [Laughter] seriously I hardly read anything except journals and science fiction because yeah what's the motivation to have robots take care of us as we age as opposed to enhancing our own cognitive abilities or um or a prosthetic body or something or cyborg why have the joy of living if you can't do anything somebody Tak care of I can't think of any advantage except that uh medicine has isn't getting you know the the age of unhandy capap people went up one year every four since the late 1940s so lifespan is so that's 60 years so people are living 15 years longer on the average than uh they did when I was born or even more than that but it's leveled off lately now I suspect that you only have to fix a dozen genes or who knows nobody really has a good estimate but you can probably double the lifespan if you could uh fix nobody knows but maybe there's just a dozen processes that that would fix a lot of things and then you could live longer without deteriorating and uh lots of people might get bored but they'll self- select I don't know what's your answer um I I feel that creating I feel that AI is more um as the goal is not to help take care of people but to compliment what we already have entertain us you could also look at them as our descendants and we um they we will have them replace us and uh just as a lot of people consider their children to be the next generation of them and I know a lot of people who don't so it's it's not Universal but what's the point of anything I don't want to get [Laughter] in we might be the only intelligent life in the universe and um in that case it's very important that we solve all our problems and make sure that something intelligent persists I think Carl San had some argument of that sort uh if you were sure that there were lots of others uh then then it wouldn't seem so important who's the new Carl Sean is there any who's the who's the no is there a public scientist who guy Isa all the oh Tyson Brian Green he's he's very good uh Tyson is the uh astrophysicist Brian Green is a great actor it's quite impressive yeah when would you say a machine has a sense of self like we think there's something that's like a self inside us because partly because there's some processes we don't like sh from us well I think that's a funny question because uh if we're programming it we can make sure that the machine has a very good uh abstract but correct model of how it works which people don't so people have a sense of self but it's only a sense of self and it's it's just plain wrong in almost every every respect and uh so it's a it's it's a really funny question because when you make a machine that really has a good useful rep representation of what it is and how it works it might be quite different have different attitudes than a person does like it might not consider itself very valuable and say could say oh I could make something that's even better than me and jump into that and so it wouldn't have the it might not have any self-protective reaction because if you could improve yourself then you don't want not to whereas we're in a state where there's nothing much we can do except try to keep living and uh we don't have any alternative if stupid thing to say I can't imagine getting tired of living but lots of people do yeah think about creative thinking as a way of thinking and where does this creativity come from or anything that comes I had a little section about that somewhere that I wrote which was the difference between artists and scientists or engineers and Engineers are in have a very nice situation because they know what they want because somebody's ordered them to make a in the last month three times I've walked away from uh my computer and how many of you have a Mac with the magnetic thing and three times I pulled it with by tripping on this and it fell to the floor and didn't break and I've had Max for 20 odd years or since 198 when did they start oo 30 years and uh they used they have the regular Jack power supply in the old days and I don't remember and usually when you pull the cord it comes out here's this cord that Steve Jobs and everybody designed very carefully so that when you pull it nothing bad would happen but it does how do you account for that it used to be better when plugs were perpendicular to and now it's kind of well it's quite a wide angle so it works at like a certain angle and the cable now instead of naturally lining that area actually naturally lies in the area where it doesn't work well what it needs is a little ramp so that it would slide out [Music] I mean it would only take a minute to file it down so so that it would slide out but they didn't I forget why I mentioned that but Engineers creativity right so what's the difference between an artist and engineer well when you do a painting it seems to me if you're already good at painting the nine10 of the problem is what should I paint so you can think of an artist as 10% skill and 90% trying to figure out what the problem is to solve whereas for the engineer uh somebody's told him what to do uh make a better cable connector and so he's going to spend 90% of his time actually solving the problem and only 10% of the time trying to decide what problem to solve so I don't see any difference between artists and Engineers except that um the artist has more problems to solve than it could possibly solve and usually ends up by picking a really dumb one like let's have a saint and Three Angels where will I put the third Angel that's the engineering part just just improvising so the to me the media lab makes sense the artists and or semi artists and the scientists are doing almost the same thing and if you look at the more Arty people they're a little more concerned with human social relations and this and that and others are more concerned with very technical specific aspects of signal processing or semantic representations so on but so I don't see much difference between the Arts and the sciences and then of course the Great Moments are when you run into people like Leonardo and Michelangelo who uh get some idea that requires a great new technical Innovation that nobody has ever done and it's hard to separate them I think there's some place where Leonardo uh realizes that the lens in the eye would mean that the image is upside down on the retina and he couldn't stand that so there's a diagram he has where the corny is curved enough to invert the image and then the lens inverts it back again which is contrary to fact but he has a sketch showing that he was worried about if if the image were upside down on the retina wouldn't things look upside down but yeah I have question did you ever heard about high high temporal memory and like temporal temporal memory like there's a system that great that try toate how our brain reords stuff they have a company called grandma and they going to release a at the end of this year on it and there's like some research they have paper well I'm not sure what this is Jeff Hawkins project I don't yeah I haven't heard about 10 years ago he said Hawkins yeah yeah well he he was talking about 10 years ago how great it was and I haven't heard a word of any progress is there some have anybody heard of there's a couple of books about it but I've never seen any claim of that it works they wrote a ferocious review of The Society of Mind in which came out in 1986 and uh the Hawkins group existed then and had this talk about a hierarchical memory system but you have any as far as I can tell it's all Bluff nothing happened I've never seen a report that they have a machine which solved the problem but let me know if you find one because uh oh well Hawkins got really mad at me for pointing this out but I was really mad at him for having four of his assistants write a bad book review of my book so I hope we were even uh if anybody can find out whether I forget what it's called you remember its name was the name of this company uhhuh well let's find out if it can do anything yet Hawkins is wealthy enough to support it for a long time so it should be good by now yes going to solve a problem people first start out with some sort of um uh classification in their head kind of problem it is or is that not necessary yes that's uh well there's this huge book called human problem solving which was I don't know how many of you know the names of new and Simon originally was new Shaw and Simon and in the believe it or not in the late 19 50s they did some of the first really productive AI research and uh then I think in 1970 so that's uh sort of after 12 years of uh discovering interesting things their main Discovery was the gadget that they called GPS which is not global positioning satellite but General Problem Solver and uh you can look it up in the index of my book and there's a sort of one or two page description but if you ever get some spare time um search the web for their early early paper by Newell and Simon on how GPS worked because it's really fascinating what it did is it looked at a problem and found some features of it and then looked up in a table saying that if there's this difference between what you have and what you want use such and such a method so it was sort what I called it I renamed it A Difference Engine as a sort of joke because the first computer in history was uh the one called The Difference Engine but it was for predicting tides and things anyway uh they did some beautiful work and there's this B book which I think is about 1970 called human problem solving and what they did is uh got some people to solve problems and they trained the people to talk while they're solving the problem so some of them were little cryptograms like uh if each letter stands for for a digit I've forgotten it pat do you remember the name one of those problems John plus Joe John plus Jane equals Robert or something that I'm sure that has no solution but uh those are called Crypt arithmetic and uh so they had dozens or hundreds of people who would be trying to talk aloud while they're solving little puzzles like that and then uh what they did was uh look at exactly what the people said and how long they took and uh in some cases where they moved their eyes they had an eye tracking machine and then they wrote programs uh that showed how this guy solved a couple of these cryptarithmetic problems then they ran the program on a new one and in some rare cases it actually solved the uh the other problem so this is a book which uh looks at human behavior and makes a theory of what it's doing and the output is a rule-based system so it's uh it's not a very exciting Theory but but uh there had never been anything like it in py you know it was like Pavlov discovering conditioned reflexes for rats or dogs and nle and Simon are discovering some rather higher level um almost a Rodney Brooks likee system for how humans solve some problems that most people find pretty hard see um anyway uh what there hasn't been is much I don't know of any followup they spent years perfecting those experiments and writing about results anybody know anything like that what psychologists are trying to make real models of real people solving Jo problems it has a green light it has a green light but the switch was up boo oh it yes did that particular study um try and see when a person gave up on a particular problem solving method how they switch to another which one they switch to based on it has it has inexplicable points at which the person suddenly gives up on that representation and he says oh well well I guess I guess R must be three did I erase well yes it's got episodes and they can't account for the these little jerks in the script where the model changes and sorry and they announce those to be Mysteries and say Here's a place where the person has decided the strategy isn't working and starts over or it's changing something the amazing part is that their model sometimes fits what the person says for 50 or even a 100 steps the guy is saying oh I think two must be Z must be two and P must be seven and that means p+ Z is nine and I wonder what's nine and uh so their model fits for very long uh strings maybe two minutes of the person mumbling uh to themselves and then it breaks and then there's another sequence so new actually spent more than a year after doing it verbally uh at tracking the person's eye motions and trying to correlate the person's eye motions with what the person was talking about and guess what none it was almost as though you look at something and then to think about it you look away it was uh new was quite distressed because he spent about a year crawling over this data trying to figure out what kinds of mental events caused the eyes to change what they were looking at but when the problem got hard you would look at a blank part of the thing more more often than uh the place where the problem turned up so conclusion it that didn't work when I was a very young student in college I had a friend named Marcus singer who was trying to figure out how the nerve in the forelimb of a frog worked and so he was operating on tadpoles and he spent about six weeks prep moving this sciatic nerve from the leg up to the arm of this tadpole and then they all got some fungus and died so I said what are you going to do and he said well I guess I'll have to do it again and I switch from biology to [Laughter] mathematics but in fact he discovered the growth hormone that he thought came from the nerve and made the if you cut off the limb Bud of a tadpole it'll grow another one and grow a whole it was a n I'm sorry it's salamander it'll grow a new hand if you wait till it's got a substantial hand it won't grow a new one but uh he discovered the hormone that makes it do that yeah oh one of the um questions from the home that kind of relates to problem solving thing um a common theme is having multiple ways to react to the same problem but how do we choose which options to add as possible reactions to the same problem oh so we have a whole lot of if then and we have to choose which if I don't think I have a good theory of that yes if you have a huge rule-based system and they're what does Randy Davis do what if you have a rule-based system and a whole lot of rules fit ifs fit the condition do you just take the one that's most often worked or if nothing seems to be working do you certainly don't want to keep trying the same one I think I mentioned Doug lett's rule some people will assign probabilities to things to uh behaviors and then uh pick the way to react in proportional to the probability that that thing has worked in the past and Doug lenet thought of doing that but instead he just put the things in a list and whenever a hypothesis worked better than another one he would raise it push it toward the front of the list and then whenever there was a choice it would pick if all the rules that fit it would pick the one at the top of the list and if that didn't work it would get demoted so that's when I became an anti- probability person that is if just sorting the things on a list worked pretty well uh our probability is going to do much better no because if you do probability matching you're worse off than than what Ray solomonov discovered that if you have a set of probabilities that something will work and you have no memory so that each time you come and uh try the I think I mentioned that the other day but it's worth emphasizing because nobody in the world seems to know it suppose you have a list of things P equals this or that or that and uh in other words suppose there's a 100 boxes here and one of them has a gold brick in it and the others don't and so for each box suppose the probability is 0.9 that this one has the gold brick and this one has 0.01 and this has [Music] 0.01 let's see how many of them so there's 10 of these that makes now what should you do suppose you're allowed to uh keep choosing a box and you want to get your gold brick as soon as possible what's the smart thing to do should you but you have no memory or maybe the gold brick is decreasing in value I don't care but so should you keep trying 0. n if you have no m memory well of course not because if you don't get it the first time you'll never get it whereas if you tried them at random each time then you'd have 09 chance of getting it so in in two trials you'd have what am I saying in a 100 trials you're pretty sure to get it but in e00 Trials uh almost certain so if you don't have any memory then probability matching is not a good idea certainly picking the highest probability is not a good idea because if you don't get it the first trial you'll never get it if you keep using the probabilities um at what am I saying anyway what do you think is the best thing to do it's to take the square roots of those probabilities and then divide them by the sum of the square roots so it adds up to one and uh so a lot of psychologists design experiments until they get the rat to match the probability and then they publish it sort of like the uh but if the animal is optimal and doesn't have much memory then it shouldn't match the probability of the unknown it should uh end of story every now and then I search uh every few years to see if anybody has noticed this thing which uh and I've never found it on the web yeah so earlier in the course class you mentioned that their rule Bas methods didn't work and that several other methods were tried in the 20160 80s could you go a bit about what these other methods were that have well I didn't didn't don't mean to say they don't work rule-based methods are great for some kinds of problems so most uh most systems uh make money and you know if if you're trying to uh make hotel reservations and things um this business of rule-based systems uh has a nice history a couple of AI researchers really notably Ed Fen bam who was a student of new and Simon started a company for making rule-based systems and Company did pretty well for a while until uh and they maintained that only an expert in artificial intelligence uh could be really good at making rule-based systems and so they had a lot of customers and quite a bit of success for a year or two and then some people at Arthur D little said oh we can do that and they made some systems that worked fine and uh the market disappeared because it turned out that you didn't have to be good at anything in particular to make rule-based systems work so but for doing harder problems like translating from one language to another uh you really needed to have more structure and you couldn't just take the probabilities of words being in a sentence but you had to look for diagrams and trigrams and have some grammar Theory and so forth so but generally if you have a ordinary data processing problem try a rule-based system first because if you understand what's going on it's good chance you'll get things to work I'm sure that's what the uh Hawkins thing started out as I don't have any questions sure computers or machines can use relatively few electronic components to run a batch of different types of thought operations all that changes is data over the operation runs in the critic selector model are resources different bundles of data or different physical parts of the brain which model um the uh critics LE model oh actually I've never seen a I've never seen a large scale theory of how the brain connects its there doesn't seem to be a a global model anywhere anybody read any uh Neuroscience books lately I mean I I just don't know of any any big diagrams here's this wonderful behavioral diagram so how many of you have run across the word ethology just a few there's a branch of the psychology of animals which is thanks which is called ethology and it's the study of instinctive Behavior Uh so these and the most famous people in that field who well tin Bergen niiko tinbergen and Conrad lorence are the most famous I've just lost the name of the guy around the 1900 who wrote uh a lot about the behavior of ants anybody ring a bell so he was the sort of the first ethologist and these people don't study learning because it's hard to I don't know why but uh so they're studying instinctive Behavior which is what are the things that all fish do of of a certain species and uh you get these big diagrams [Applause] uh this is from a little book which you really should read called the study of instinct and it's a beautiful book and if that's not enough then there's a TW volume similar Book By Conrad Laurence who uh was a Austrian researcher um they were part they did a lot of stuff together these two people and it's full of diagrams show ing the main behaviors that they were able to observe of various uh lowcost animals uh I think I mentioned that I had some fish and I watched the fish tanks what they were doing for a very long time and came to no conclusions at all and when I finally read tinbergen and Laurence I realized I just had never occurred to me to to guess what to look for my favorite one was that whenever a fire engine went by loreno's sticklebacks the male sticklebacks would go crazy and look for a female because when the female's in heat or whatever it's called estris uh the lower abdomen turns red I think fire engines have turned yellow recently so no one I don't know what the sticklebacks do about that so if you're just in AI you really should look at at least one of these people because it's the first appearance of rule based systems in great detail in Psychology didn't get in there weren't any computers yet there must be 20 Questions left yeah while we on the topic of ethology so I know that uh early on uh people were kind of they're careful not to apply ethology to humans um until about what 60 EO Wilson with uh sociobiology so I was wondering about your opinion on that and maybe even have anecdotes about the time I pretty controversial around this area espe oh I don't know uh I sort of grew up with Ed Wilson because we had the same fellowship at Harvard for three years but he was almost never there cuz he was out in the jungle in some little telephone booth watching the birds or be or he also had a 26y old aunt aunt not aunt aunt ant I'm not sure what the controversy would have been but of course there would be humanists who would say people aren't animals but but then what the devil are they why aren't they better than they you've got to read this it's a fairly short book and you'll never see an animal as the same again because uh I swear you start to notice all these little things you're probably wrong but you start picking up little pieces of behavior and trying to figure out what what part of the Instinct system is it and lorence was particularly I think in chapter two of the emotion machine I have some quotes from these guys and uh lorence was particularly interested in in how uh animals got attached to their parents that is for those animals that do get attached to their parents like alligator babies live in the alligator's mouth uh for quite a while it's a good safe place and um lorence would uh catch birds just when they're hatching and within the first day or so some baby birds get attached to whatever large moving object is nearby and he that was often Conrad lorence rather than the Bird's mother who's supposed to be sitting on the egg when it hatches and the bird gets attached to the mother most most birds do because uh they have to stay around and get fed uh so it is said that wherever Laurence went in Vienna uh there would some ducks or whatever birds that had gotten imprinted on him would come out of the sky and land on his shoulder and um and on no one else and he has various theories of how they recognize him but but you could do that too anyway that was quite a field this thing called ethology and between 1920 and 1950 1930 I guess 1950 there were lots of people studying the behavior of animals and Ed Wilson is probably the most uh well-known successor to luren S tinbergen and I think he just wrote a book is that anybody seen it he has a huge book called sociobiology which is too heavy to read I've run out of things yes IDE had the Machinery from it um initial state of the machine be we would just start something is that dictated by the goals given to it by State I mean different agents the resources they have access to what would that state look like he's asking if you were if you made a model of the programmed to Society of Mind architecture what would you put in it to start with I never thought about that it's a great question I guess it depends whether you wanted to be a person or a marad or chicken or something are there some animals that don't learn anything must be what do the uh ones at Sydney brener study see El they learn very simple associations the little worms there was a rumor that if you fed them RNA was it them or was it some slightly higher animal it was worms what R interference is talking about yeah there was one that if you taught a worm to turn left when there was a bright light or right and put some of its RNA into another worm uh that worm would uh copy that reaction even though it hadn't been trained and this was wasn't worms that was slugs slugs I think it was yes or something a little snail likee thing uh and nobody was ever able to replicate it so it's that rumor spread around the world quite happily and uh there was was a great science fiction story trying to [Music] remember in which uh somebody got to eat some of an alien's RNA and got magical powers I think it's Larry nien who is wonderful wonderful at taking little scientific ideas and making a novel out of them and his wife Marilyn was a undergraduate here so she introduced me to Larry nien and I once got to write an artic I once gave a lecture and he wrote it up it was one of the big Thrills because nian is one of my heroes imagine writing a book with a good idea in every paragraph verer vingi and Larry nen and Frederick [Music] pole seem to be able to do that or at least on every page I don't know about every paragraph yeah to follow up on that question it seems to me that you are you almost were saying that this exist the difference between um you know these sort of animals would be the start State and depending on the start State we could either create like a chicken or a human well no I don't think that I don't think that most animals have scripts they some might but but uh you know I'd say that I don't know where most animals are but but I sort of make these six levels and I'd say that none of the animals have this top self-reflected layer except for all we know dolphins and uh chimpanzees and whatever it would be nice to know more about octopuses because uh they do so much wonderful things with their eight legs what kind of how does it manage have you seen pictures of an octopus picking up a shell and walking to some quiet place and it's got there's some movies of this on the web and then it drops the shell and climbs under it and disappears it's hard to imagine programming a robot to do that yeah so uh I've noticed both in your books and in the lecture a lot of your models and diagrams seem have a very hierarchical structure for them uh but as you've mentioned in your bu and other places uh passing between levels feedback and self reference are all very important to intelligence so I'm curious if you could discuss some of the uses of these very hierarchical models why you've represented so many things in that way and some of the limitations there well it's probably very hard to debug things that aren't so uh so we need a sort of meta Theory one thing is that for example it looks like that all neurons are almost the same now there's lots of difference in in geometric features of them but they all use the same uh one or two transmitters and uh every now and then you run across people saying oh neurons are incredibly complicated they have 100,000 connection you can find it if you just just look up neuron on the web and get these essays explaining that nobody will ever understand them because typically a neuron is connected to 100,000 others and blah blah blah so it must be something inside the neuron that figures out all this stuff uh as far as I can see it's it looks almost the opposite namely probably the neuron hasn't changed for half a billion years very much except in sort of superficial ways in which it grows because if you change any of the genes controlling its metabolism or the way it propagates impulses then uh then the animal would die in in before it was born and uh so you you can't make that's why the embryology of all mammals is almost identical you can't make a change at that level the after the first before the you can't make changes before the first uh generations of cell divisions or everything would be clobbered the architecture would be all screwed up so I suspect that the people who say well maybe the important memories of a neuron are inside it because there's so many fibers and things I bet it's it's sort of like saying the important memory in a computer is in the Arsenic and phosphorus atoms of the semiconductor so I think things have to be hierarchical in in evolution because if you're building later stuff on earlier stuff then it's very hard to make any changes in the earlier stuff so as far as I know the neurons and C anemones are almost identical to the neurons in mammals except U for the later stages of growth and the way the fibers ramify and who knows but but uh there are many people who want to find the secret of the brain in what's inside the neurons rather than outside and U be be nice you get a textbook on neurology from 50 years in the future see how much of that stuff mattered where at our time machines did you most systems have a state that they prefer to be in right like a state that they're more most comfortable in do you think the mind has such a state would it tend to certain like places something that's interesting I don't how does that apply to living things I mean this bottle would rather be here than here but I'm not sure what you mean well okay so apparently so in Professor ten bom's class he shows this like example of a of a number game he'll give you a sequence of numbers and he'll ask you to find a pattern in it so for example if you had a pattern like 10 40 50 and 55 he kind of asked the class to come up with like different things that could be describing the sequence and um you know between a choice of oh this sequence is a sequence of the multiples of five versus a sequence of the multiples of 10 or multiples of 11 um you know he says something like he phrases it like the multiples of five would have a higher prior probability but so that got me thinking like why would that be would would our minds have a preference for having as few categories as possible in trying to view the world around us like trying to categorize things in as few things as possible is what sounds very strange to me but uh certainly if you're going to generate hypothesis you have to have the way you do it depends on what what this per if what does this problem remind you of so I don't I don't see how you could make a a general if you look at the history of psychology there are so many efforts to find four laws three laws of motion like Newton's is he trying to do that I mean you're here you're talking about people with language and high level semantics and let's ask him what he meant Prof yeah this more of a social question but there's always this debate about how if AI gets to the point where it can take care of humans will it ever destroy humanity and how would do you think that's something that we should fear and if so is there some way we can prevent it if you judge by the recent by what's happened in AI since 1980 it's hard to imagine anything to fear but but uh funny you should mention that I'm just trying to organize a conference sometime next year about disasters and there's a nice book about disasters by what's his name the Royal the astronomer Royal what Martin Martin Ree um so he has a nice book which I just ordered from Amazon and it came the next day and it has about 10 disasters uh like a big meteor coming and hitting the Earth I forget the other 10 but I have it in here somewhere so I generated another list of 10 to go with it and so there are lots of bad things that could happen but I think right now that's not on the top of the list of dis disasters eventually uh some hacker ought to be able to stop the uh net firm working because it's not very secure and uh while you're at it you could probably knock out all of the uh navigation satellites and uh maybe set off a few nuclear reactors but I don't think AI is the principal thing to worry about but it should very suddenly get to be a problem and there are lots of good science fiction stories my favorite is the Colossus series by DF Jones anybody know there was a movie called The Forbidden project and it's about somebody who builds an AI and it's trained to do some learning and uh it's also the early days of the web and it starts talking to another computer in Russia and suddenly it gets faster and faster and takes over all the computers in the world and gets control of all the missiles because they're linked to the network and it says we will I will destroy all the cities in the world unless you clear off some Island and start building the following machine I think it's Sardinia or someplace so they get bulldozers and and it starts building another machine which it calls Colossus 2 and they ask what's it going to do and Colossus says well you see I have detected that there's a really bad AI out in space and it's coming this way and I have to make myself smarter than it really quick anyway um see if you can order the sequel to Colossus that's that's the second volume where the Invader actually arrives and I forget what happens and then there's a third one which was an anticlimax because I guess DF Jones couldn't think of anything worse that could happen but but but Martin Reese can yeah um going back to her question about the example and if our mind has a state that a prefers to be in would that example be more of a pattern recognition so instead of 10 40 50 55 what if it was linguistical like good fine great and you have to come up with a word that um that could potentially fit in that pattern and then that pattern could be waste answering how are you let's do an experiment how many of you have a resting state sometimes when I have nothing else to do I try to think of Twinkle Twinkle Little Star happening with the second one starting in the second measure and then the third one starts up the third measure and when that happens I start losing the first one and ever since I was a baby uh when I have nothing else to do which is almost never uh I try to think of three versions of the same tune at once and usually fail what do you do when you have nothing else to do any volunteers what's yours try not to think anything at all see how long it'll go you tried not to or to not to isn't that a sort of uh Buddhist thing yes sir do you ever succeed how do you get out of it you have to you have to think well enough of this nothingness if you succeeded wouldn't you be dead we stuck eventually some simulus will appear that is too interesting to ignore right and the threshold goes down till right even the most boring thing is fascinating yeah make a good short story yep there was actually a a movie that really got to me when I was little these aliens were trying to infiltrate people's brains and like their thoughts and to keep the aliens from infiltrating your thoughts you had to think of a wall which like didn't make any sense at all but but now but now whenever I try to think of nothing I just end up thinking of a wall there are these awful psychoses and about every about every five years I get an email from someone who says that please help me there are some people who are putting these terrible ideas in my head have you ever gotten one Past and uh they're sort of scary because the you realize that maybe the person will suddenly figure out that it's you who's doing it if [Music] they of them together once I think they married I remember there was once one of them came to visit actually showed up and he came to visit Norbert weiner who was famous for I mean he's the cybernetics person of the world that was and this person came in and he got between weer and the door and started explaining that somebody was putting dirty words in his head and making the grass on their lawn die and he was sure it was someone in the government and uh this was getting pretty scary and I was near the door so I went and got lein it's a true story was nearby and I got lein to come in and lein actually talk this guy down and took him by the arm and went somewhere and I don't know what happened but weer was really scared because the guy kept keeping him from going out lein was big reer is not very big anyway uh that keeps happening every every few years I get one and I don't answer them it's probably doesn't he's probably sending it to several people and uh I'm sure one of them is much better at it than we are how many of you have ever had to deal with a obsessed person how did they find you I don't know I found a number of people in the media lab actually don't answer anything but if they actually come then it's not clear what to do last question thanks for coming