Thinking Fast and Slow | Daniel Kahneman

[Music] good evening I'm Alexander Rose the director here at long now we don't have a long short tonight but I thought I would take this opportunity to go over a little bit about where we've come from it turns out that that this fall will be our tenth year of doing these seminars and actually could could we get the house lights up just a little bit house lights a little bit cuz I actually want to see this how many of you were at the very first Brian you know seminar 10 years ago well about maybe about 50 nice not bad so as Paul Sappho who's here tonight knows on our board has said this is his favorite conference in that it's just been going for 10 years basically it's just a very long drawn-out single conference so this fall we'll be hitting somewhere around 120 talks and about 7500 people come to them annually like tonight but one thing that I think a lot of the people who come to these don't understand or don't see is the other half of our audience in fact the other 99% of our audience we get about 500,000 downloads or views of this through our podcast and the video every year so it's by far our largest outreach program with long now the the way that we pay for this has largely been through all of you members so how many members are here tonight so as you know you don't purchase tickets to come in every night but you do pay a monthly membership and that's what that's what keeps this program going we also have a series of individual sponsors most of which are here tonight as well who also pay for that and that basically makes between those two things that's over half of long-nosed operation budget so thank you to all of you our sponsors and our members for making that happen I really appreciate it [Applause] so that the real question is what do we you know where do we go from here ten years and five hundred thousand people at least listening and watching annually and one of the main questions we often get is you know why don't you do X speaker who you know fill in the blank and these talks cost us somewhere between twelve and twenty thousand dollars to put on even with amazingly generous speakers who don't charge us to do the their normal speaking fee or any speaking fee to come so we can't just do every speaker but what we are now embarking on doing is making our salon project which will allow us to do smaller events and much more impromptu events but we will still be able to webcast them and and all of that we are a little bit past halfway in our fundraising for that and we have just started demolition this week and we've started and we'll be starting construction in the next couple weeks and so thank you all for getting us to halfway and if any of you are interested in helping to get us to the second half way we are we would be very appreciative of that as well and the last thing I wanted to say is that this whole thing is done with an amazingly small staff I think there's at any given day there's only about four or five of us in the office and these events are expertly engineered and put together by daniele Engleman and her crew and I really want to thank them for doing that thank you very much good good evening I'm Stuart brand from the long now foundation the idea of the salon conversation plus fluids coffee tea alcohol all of them exotic and full of time in various respects there's a Bristlecone gin that we're going to be serving among other things our speaker sold this book to me three times because I'm listening to it in the car and then I go get the printed Edition and Mark the parts I really like and then the Kindle version I take on the plane and I come back from that and I mark the parts in the printed version that I really like and I have to do that because practically every problem I'm working on in various respects including bringing back extinct species has solutions buried in what our speaker has gotten into in his lifetime of psychological research and we say the long now foundation that one way we state the goal is to make long-term thinking automatic and common instead of difficult and rare well I think we can add to that we would like to make slow thinking but our speaker calls system to thinking automatic and common instead of difficult and rare here to talk about it is Daniel Kahneman [Applause] [Music] [Applause] well it's a pleasure to be here it's a bit of a surprise actually in some way I you know that my book Thinking Fast and Slow came out more than a year and a half now and I'm sick and tired of talking about it and but when stood brand asks you to come in and do anything I suppose I mean he hasn't asked me to do many things but it's I think it must be very difficult to refuse anything that Stuart wants you to do and also I was particularly interested because of the tremendous contrast between our temperaments the long now foundation as I perceive it is really you know it's it could stand as a symbol for optimism not you know long range optimism I mean thinking big and thinking long and thinking for and and I am really a pessimist train fundamentally a pessimist I don't you know I think you know I don't want to think about the future because I don't think it's going to be too good I attend I take some pleasure in the fact that I'm very rarely disappointed by anything I mean which is you know one of one of the major advantages I think of being a pessimist but and this was what brought me here I mean I'm I'm going to tell you how I think and so I'm going to expose you to a semester thinking for which I suppose must be in real supply here but but I really longed to be convinced I would like to be convinced okay so let me tell you about fast and slow thinking you know that there are two fundamental ways of thinking or two fundamental ways in which thoughts come to our minds is completely obvious so it's so obvious actually that not everybody sees it immediately and psychologists haven't done it you know haven't put a lot of work into that basic idea but the distinction is very clear you know so if you say 2+2 something happens to you and and if you say 17 times 24 I always use the same two examples probably not much happens to you as so certainly you don't 408 doesn't come to mind if you unless you have misspent your youth or something what would so there is that difference if you want to get to 408 if you want to know the answer to 17 times 24 you've got to recognize versus a multiplication problem you have to bring up the program that you learned in school you've got to apply it this is this is work it is mental work and and it's very easy to notice that it's mental work because there are physiological changes that are associated with it it's completely different as as a mental event from what happens when you know I mentioned 2+2 and the number 4 sort of pops into your mind you didn't try to bring it to mind you didn't want to bring it to mind particularly it just happened so the difference between fast thinking and slow thinking is that first thing is something that happens to you primarily and slow thinking something that you do this is it has elements of intention it has elements of attention it has elements of control and that's the key distinction now we when we apply slow thinking when counter limitations and then that's a very important between fast and slow thinking effortful thinking is something is our ability to apply effort as quite limited and as we really cannot do too many things at once if they demand effort we are likely to fail if we try to do too many things at once so you can compute 17 times 24 you cannot do it while making a left turn into traffic and you shouldn't try so that's the and you shouldn't try because you'll fail and in one of the other the two tasks and and one of them is much worse than the other normally by the way we do have a mechanism that prioritizes so that mostly when we have competition between two tasks and we have to choose some you know we something has to choose some and we actually have a pretty good idea of where that thing that chooses is is in the brain a choice has to be made and we're very good at that usually not always but usually we're very good at that now we need effortful thinking to do many things we probably need it whenever we want to hold two ideas at once and and compare them we can as you'll see we can easily handle a single idea or a single very complex idea single interpretation but holding two possibilities at once like you do when you do a deliberate choice that seems to require effort self-control which is not slow thinking but the characteristic of self-control is that it two demands effort so that if you are very busy or very tired your self-control is to some extent depleted and we know that from many experiments but you know to give you an example the way that psychologists sort of preempt the work of system two or makes make the way that we make people as we say cognitively busy is we give them a set of digits to keep in their mind in their head and and later we'll test them and while they are trying to not to forget those seven digits which require some rehearsal and some mental work we give them other tasks and it turns out that when you are holding seven digits in your head in very obvious ways your self-control is impaired so for example people use more sexist language and when their self-control is impaired they are they the choices between desserts is somewhat of influence they are more likely to pick the rich and luscious chocolate cake and less likely to go for the more virtuous food salad that's because normally we control ourselves and when our ability to control ourselves is is impaired we do what comes more naturally we yield to temptation we normally think of ourselves as system - we're identified with what we think consciously we are identified with the things that we choose consciously the when you generate a thought like computing 17 times 24 you are the author of that thought this I mean we feel some pride in things that just occurred to us and happened to us but the authorship of a thought the idea that it was produced by deliberately that is a marker system - and we I think many of us have a sense which I believe is erroneous that when we have a thought that just came naturally there is reasoning behind it that as we could produce the reasons that led to this thought this is I think an illusion that many of us have in many cases as I'll try to show you there is no reasoning behind the thought there is another process there is a process that they'll call an associative process in memory so let me give you another example of system 2 and how it works and how it fails and that's a puzzle anybody who's read the book knows that puzzle it's the best single example of its kind and and the puzzle is very straightforward it's a bat and a ball costs $2 10 the bat costs a dollar more than the ball how much does a ball cost and the beauty of that example is that an association comes to mind really I think everybody has a dissociation and that's 10 cents the number ten cents pops up now that number is wrong and and it's very easy to find out that it's wrong to them convince yourself that it's wrong because you know if it's 10 cents for the ball then it's a dollar and ten cents for the bat and then it's a dollar 20 it's totals or something is wrong the answer is five cents what's interesting about this problem is that 50 percent of students at Harvard fail it and they fail it when it's given in a written questionnaire so what happens to those 50 percent of people who fail that test what happens is something really quite interesting they had that association ten cents and they didn't check that as we know about everyone who makes a mistake on that problem they did not check themselves an idea came to mind and they endorsed it and that idea of endorsing thoughts that come to mind spontaneously this is what system two does in many situations I would argue in most situations it endorses ideas that came from the associative system which is a system over which we have no control so this is a case you know that particular puzzle it there's a really broad hint that you should check because the problem is so obvious that if you give the obvious answer you know it's very lazy actually to give the obvious answer and that's a characterization I apply to system two it tends to be lazy it tends to operate by law of least effort there are individual differences some people are much lazier than other people in in the way they think but but all of us the law of least effort is really a fundamental psychological principle that is when when there are alternative ways of doing something over time we tend to gravitate for the way that minimizes F there is a calculation that is the value of actions there is a calculation that's performed in the brain and effort is a cost in that calculation and so we tend to minimize effort as I said some people are more minimized effort but more than others so that's the way that I'm going to describe how we think a lot of the time the basic mechanism in that suggested answer pops up from the associative system from fast-thinking from what I call system 1 thinking and in most cases the suggestion is endorsed and we believe what we see we believe what we think so that's now that suggests that system 1 is the very large extent in charge because system 1 generates those possibilities those suggestions that in general system 2 will adopt and accept now system 2 has control to some extent that as we could trick people at least in that problem people have the opportunity to trick themselves and they have the opportunity to decide not to accept that suggestion but actually to do something else but much of the time it is system 1 that actually runs the show so that's the description I'm trying to give you I say conclude very general way of how the mind works we have an associative system that generates automatically I mean you know without intention without volition generates ideas generates perceptions generates an interpretation of the world quite broadly and then we have sitting if you will on top of that we have a system that monitors the quality of you know our behavior we self monitor we self control but that system is relatively it's very costly to apply and much of the time the monitoring is very light much of the time though monitoring is also unnecessary and that is because that associative system is an absolutely magnificent piece of work so most of the time what we do naturally what we might say you do it what we think intuitively all of that tends to be right most of what we do is just fine and intervention by system to would just slow things up and make them worse and you know all athletes know that very well that when they when they start thinking about what they're doing it disrupts the quality of of what they do because their skill is disrupted by the operational system too so I'll talk mostly about system one and its characteristics and I'm not going to describe it in full but I want to give you a sense for how it works so the automatic operations of system 1 are involved for example when well you know if I mention the capital of Russia that's like to pursue that something is retrieved automatically from memory when you recognize a friend the name might pop up but actually as you can see reckon recognising friend is much more than retrieving a name you're retrieving a whole lot of things are happening to you it's a whole network of ideas associations emotions all of which are connected with that friend and all of which are activated if your friend looks sad or tired there will be something that's equivalent to search although it's search is a little too deliberate - but what will happen is in memory if there is a memory a causal explanation if your friend looks sad and good and tired and and you there is something that you know that could explain it that will come to mind automatically if you had heard that that his wife is ill that will come to mind as an explanation of why you are struck by the fact that he looks sad or tired when I say your mother a lot happens including an emotional response and something also happened to your face probably that is associated with that emotion that came to your mind and it would be sad or happy angry or loving but what happens is not a single thought it's a lot is happening including you know emotional manifestation facial manifestations and so on now some of the responses of system one are probably universal and unique so obvious example is we don't learn I mean we are wired to learn to see things and seeing as I just interpreting the visual world is the prime example for me of an operation of system 1 it is automatic it is fast it is in general extremely accurate and those are characteristics of the way that system works now there are other things that are clearly innate or at least universal to give you an example of an association you know so consider the two words you know one classic psychological example one is maluma and the other is Steketee and if I asked you to which word go with which gesture you know this gesture or this restroom it's completely obvious what goes with what so that seems to be universal it seems to me cross culturally universal according to some people have done research on that that's built in fear of spiders is probably built in and I want to tell you personal experience which I think that's a very unusual experience and that happened to me of something that is built in an inn system one and it's a bit of an embarrassing story but a few years ago we have a house on grizzly peak in Berkeley which is very nice Road on top of the hill and and I was thinking a walk late in the evening and really quite late it was deserted and dark and a car came on the opposite direction and and they were rowdy youth were in that car they you know they they really sounded they sounded as if they'd had too much beer I mean they they really didn't sound very nice and I I had that I have that impression as they were whizzing past in the other in the opposite direction and then they stopped and that there was no obvious reason for them to stop and then a few of them got out of the car I could hear that and they started walking quite fast in warn me and it was unclear you know what when they wanted but it really didn't sound friendly and and they were coming a bit closer and then something happened to me which took me it took them by surprise it took me by surprise I i reenact and I don't know what I did with my hands but I roared at the top of my voice I mean like where that came from no idea absolutely no idea but it came from way down deep we have you know there is an evolutionary history there are things like fear of spiders and and you know how how threatened animal we reacts to a threat and you know possibly how an old monkey reacts to threatening juveniles that is built into system one but of course most of what is in system 1 in associative memory is learned and what we have is detailed knowledge of the world that we can access and you know what in terms of you know for a computer SFO in the computer age it may not be all that that fast but it's pretty extraordinary and you know I have an example that I mentioned in the book it's the best and it's very hard when you've given lots of book talks to think of better examples so the example I have in the book is of people listening to sentences and acts and while their brain events are recorded and a sentence spoken in an upper class British male voice which I will not attempt to to simulate here it says I have launched the tools all down my back and within about the third of a second there is a characteristic response in the brain which indicates surprise and if you stop to think about it this is astonishing that is what's involved you know and that is all system one it's completely automatic it is not there is no intention that's the way memory works you had to somehow detect and recognize this is an upper class you know British male voice activate a stereotype of what upper class British like relate that to the question of whether they have lodged the twos down their back that doesn't fit the stereotype and there's a surprise and interestingly enough the reaction to that surprise is system two is activated and is when there is surprise we are mobilised to perform antal effort to perform mental work to make sense of things normally when things go routinely and are not surprising system one sort of system 1 thinking flows when there is a surprise we can catch the mobilisation of effort skills are automatic so driving is an automatic skill chess playing at a high level is an automatic skill that is you know tres players master tres players every single move that occurs to them is a strong move and as they they select among strong move and they knew that consciously and they you know they can compute for for a while but the basic selection of moves is is performed completely automatically so what happens is when we acquire skills an activity that at first you know this is the way we all learn to drive an activity that at first requires deliberate thinking and effort where where is the accelerator where is the brake all that eventually becomes completely automatic and and we know how that happens we know our skills are required we know that it takes a lot of practice and we know that immediate feedback is important and and ultimately all of that is goes into system one so we have a rich representation of the world a representation of skills and an ancestral memory of you know the time that I mentioned earlier now there are interesting mixtures I think there is evidence that recognizing that somebody's afraid maybe innate but recognizing somebody is styled and doesn't look well I would expect as a learned skill I mean I remember being very puzzled as a child you know how did my mother know that you know somebody looks tired but eventually I worked it out and there are cues we're not not aware of accuse but the categorization of people is looking well looking sick looking time that happens automatically it's a learned skill now how does this whole thing works I've mentioned the word associative memory and it covers a lot of ground because it covers that what happens when there is a stimulus how does a response including a mental response get generated all that is in associative memory we have a primitive notion of what associations are like you know night day or bread butter but that's not the interesting associations they are much much richer than that if if I mention the relationship between China and Japan now something happens to you when I mentioned that and it wasn't a single idea it wasn't a single word it was probably you could feel it the whole network of ideas a whole concept very rich actually network of ideas was activated now how do we know that it's activated we know because the activation of you know what happened to you when I mentioned the relationship between China and Japan we know that you're now prepared for many other events that could happen and there is a very simple test of that and the test is to present whispered words and to see they've presented at a very so that they're barely audible and see which words are recognized in which are not and after when I've mentioned the relationship between China and Japan you will be quicker to recognize the word island and there were Navy and possibly the world war and you are prepared for a lot of things that belong in the network of ideas that were activated by that concept now you don't have the whole history but but you can feel that this activation was quite dreadful and quite rich now here I'm coming to something that's really quite important which is that how we interpret events and stimuli as they occur is again tends to be determined by the context by how it fits with other things that are going on in our mind at the time so stand an example not mine it's standard psychological example is an approach the bank when I mentioned that sentence the word that sentence is really ambiguous you probably thought of the bank as a financial institution but if it was in the context of fishing the bay and approach the bank would have a completely different interpretation what's important and interesting about this is that you're not really aware of the interpretation that was rejected the choice was made the choice is context dependent and you're not aware of it oh I have a story again a personal story to tell about that particular mechanism it's one of the best examples it happened after I finished the book otherwise I would have had to ask my wife's permission to include it because she's involved we were out to dinner with friends and and then you know we were talking about about that evening and my wife said of a man she said he is sexy all right which and and then the next thing that she said was truly bizarre and in fact it was it was very very odd because she said he doesn't underestimate himself and I thought that's a very very puzzling remark and I asked know what what are you saying you know what's what's going on it turned out but she had actually said I'm a little hard of hearing she had actually said he doesn't underestimate himself this this is a rich example because because it never occurred to me and that I think is something that you may recognize in yourselves as well it never occurred to me to think that if this sentence was so bizarre probably I misheard it that didn't look it to me I knew what she had said you know the only question was why did she say such a thing and that is a very common occurrence we take we accept you know what we see and then we look for any interpretation we do not question the interpretation in many cases that turns out to be a fundamental characteristic of the way our mind works is the fact that you know we hear the word and we are not aware of alternative interpretations this is true in perception as well when we present and big us stimuli an interpretation is chosen it's chosen by the mechanism we're not aware of the choice and we're not aware of the rejected alternatives so what is characteristic of the way associative memory works is that it produces an interpretation of the world that is coherent and I want to elaborate a bit on what that word coherence means because there are really two very different kinds of coherence there is a logical coherence and logical coherence has to do with operations of reasoning so that you know if you believe that if you believe that if a then B and if you believe that a then you must believe that be you know that's anything else would be incoherent in that sense so logic is a system of coherence this is not the kind of coherence that system one or fast-thinking or associative memory the the relevant concept for associative memory is fit rather than logic it's things that go together and as in the example I gave you earlier maluma goes with a certain kind of gesture and Tahiti goes with another kind of gesture that's fit that is coherence there's no logical connection there is no you know there is no logical necessity that relates it to but in terms of fit it's very obvious and it turns out that a lot of our thinking and our emotional reactions are governed by this kind of associative coherence in ways that are really quite difficult to control the psychologist Paul Rosen is an expert in making people psychologically uncomfortable and and here is something that he does very simple experiment so you take a bunch of students and there's a glass of orange juice in front of them and snicker that they can write on and he asked them to write the word cyanide in capital letters and then to put the sticker on the glass and then to drink the orange juice and and they don't like it now I mean it gets worse because he also has the example of giving people a comb and in its plastic wrapping and asking them to stir milk with a comb perfectly clean never been used in that plastic right and and drink the milk absolutely not now what happens here is there is a connection and there is an associative connection and it's associative it's emotional in the the orange juice is tainted by the label of cyanide and in a way you just don't want to come near it that's this is the way that system one works now that's kind of emotional and associative coherence is I think extraordinarily important because it explains our deepest beliefs you know when when you look around you at what people believe you know there are religious positions there they really just faithfully political positions they're really very few of them and yet and they think you know actually they think that people on the other side are sort of dumb low or ridiculous so how could anyone believe in nonsense like that that's the other people's belief where does this come from where does the certainty that we have in our own beliefs come from why is it so difficult to change people's minds on issues of faith or political opinions I mean you know you can debate things until you're blue in the face and basically very little happens except people getting angry at other people's stupidity what seems to happen is related I think to associative an emotional coherence our beliefs an association with people that we like and love and trust that's what they believe it's very difficult to find any other reason or any other mechanism that would explain why people grow into one religion rather than another and grow into a religion that they hold with complete confidence because people know what the truth is and that sense of knowing but the truth is must come from somewhere and it's rooted in a in an emotional and associative history of believing people trusting people people telling you and you're finding yourself believing in what they believe and say there are other manifestations of associative coherence one of them is what psychologists call the halo effect and it's it's easy to demonstrate by mites opposite for example the idea that Hitler adult faith love children and dogs which is true by the way and he was very kind to children and dogs that is an unpleasant idea it doesn't fit and we tend to our system tends to create a coherent image of people so the people that we like if they have one good property the rest of their properties tend to be good if we dislike them we dislike the whole thing you know people's attitudes say to President Obama so if if you like his politics very likely you like his voice if you don't like his politics very likely you think he is very big ears and and you know he has a nice voice and he doesn't big ears but the the way that we perceive things is associatively and emotionally coherent and that colors the way that we think about many things and one of the almost frightening examples of that I think is what is called the affect heuristic which is that people's attitude to things is people's beliefs about things are determined by whether they like them or not and I can give you an example of that um Paul slovic the psychologist who is introduced this label of the effector istic is interested in people's attitudes to technologies and there are technologies that people like and there are technologies that they dislike and you know so there are differences in taste and and a very interesting thing happens that when you collect people's beliefs above the various technologies if there is a technology that they don't like then it has a lot of costs and virtually no benefits so the correlation between the number of costs that people can think of in the number of benefits that people can think of is sharply negative in the real world that's very unlikely to happen that way in general choices are difficult because there is a positive correlation between costs and benefits but in people's mind that's not the way it works now you take new people with a particular technology called you know it could be fracking or it could be wind turbines and and you provide people new information about costs that they hadn't thought of before so they now like the technology less and interestingly enough their ability to think of benefits of the technology is impaired that is you haven't said anything about benefits all you've told them there are costs that you didn't consider and now benefits don't come to mind because we tend to create an interpretation of the world that is emotionally and associatively coherent here is another manifestation of the way the associative system works and it's a question and the question is is the following logical argument valid that is does the conclusion follow from the premises and the argument is all roses or flowers some flowers faint quickly therefore some roses fade quickly in this valid argument or not now the argument is not valid because you know it could be that the none of the flowers that fade quickly is a rose so Rose don't there are no roses let's read quickly but about 80 percent I think above 80% of students believe the state the statement is logically valid the argument is logically valid what happened well what happened is the consequence is true for people believe the consequence that makes them believe the argument that favors the consequence this is not the way that thinking is supposed to work thinking's supposed to work from premises to conclusions but in fact what happens is that if you believe in the conclusion you tend to believe in the arguments that go with the conclusion this goes a fairly long way to explaining political beliefs you believe in the conclusions and you are inclined to believe in any argument stronger weak that is compatible with your beliefs by the way it's a the same logical structure it's trivially easy to see to see it when it's presented in terms of algebra then you see the Venn diagrams very easily or if the conclusion is false that you obviously see that the argument is not valid so there is a really important reason for lesson from this that people reason backwards and there is another psychological reason that I want to develop an EM coming close to the end of the story and this is the following notice would happen in that puzzle you were asked so students are asked one question which is is the argument valid they apparently answer a different question first and the different question is is the conclusion true and they infer validity from true and they are not absolutely not aware that they are doing this they answer a different question from the question they were asked this it turns out is a very general mechanism now let me say something in conclusion think about what it means for a mind to generate interpretations of a world that are characterized by associative and emotional coherence well that means that the individual went this mine lives in a world that is a lot simpler than the real world because the real world doesn't have that kind of coherence Hitler did like dogs and little children life is more complex than that it is more complex than we're inclined to see and we're not geared to perceive the complexity now following on the other line here of how did we answer the question about the valid argument that we answered a difficult question by answering a simpler one this it turns out is a very general mechanism system one comes up with answer system one is really stumped it comes up with answers to questions that it shouldn't be able to answer and I call that substitution that you substituted an easy question for the hard one and it really happens a great deal so let me give you an example one of the most striking experimental examples that I'm aware of this was during the period of well in the 90s I think that experiment was done there was a fair amount of terrorism in Europe and Americans were worried about traveling to Europe because of terrorism an experiment was done basically about travel insurance and how much would be people be willing to pay for travel insurance and some people are asked to put a price how much they would be willing to pay for a policy that pays $100,000 in case of death during your trip for any reason other people were asked how much would you pay for a policy that pays $100,000 in case of death in a terrorist incident during this trip now the second policy is worth a lot more than the first now that really doesn't make sense because if you you know dying in a terrorist incident is not the only way of dying in the trip in fact it's quite unlikely so logically there is no question about witches worth more but that's not the way that people think notice they are not asked to compare the two policies if they were us to compare the two policies many people would figure it out that one of them is worth is worth more than the other but when they see one policy at a time so you can't apply logic system two cannot be applied to solve that problem to detect a logical connection then we know what people do they answer in terms of how frightened they are how afraid they are and people are more afraid of dying in a terrorist incident than they're afraid of the and that actually and that makes perfect sense I mean in terms of an association you have more vivid images they're more frightening and and if you decide on the value of insurance by answering the question how afraid am I at this event then you will pay more for insurance against terrorist incident than for insurance against death many of the examples know I did the research on judgment my book is dedicated to my late colleague Amos Tversky and we started well more than 40 years ago now and we dealt with mistakes that people make in statistical judgments or failures to apply statistical gentlemen and one of the examples that we had was about a fellow named Steve and we said he had been sampled at random from the population now and and we described Steve as a meek and tidy soul with a passion for detail relatively little interest in people and then we asked the steel more likely to be a farmer or a librarian and the answer is immediate most people immediately pick up that Steve looks like a stereotypical librarian I don't know the stereotype is true I doubt it but but that's what they that's what people do and they think that Steve is more likely to be a librarian now in fact people also if they stop to think about it would realize that in the United States for example there are about twenty times as many male farmers as male librarians there aren't many farmers and there are many librarians many more male farmers Steve is actually as I described it there are more meek and tidy Souls on farms they were than there are in libraries that doesn't come to mind we answer by comparing Steve to a stereotype and Steve is more like the stereotype of librarians then Steve is like the stereotype of farmers I will give you one more example just to show you how thoughts that we generate are dominated by associative thinking and illustrate another characteristic of associative thinking as I do this and this is an example which I discuss in the book in some detail actually it's it's about the girl I call her Julie and she is a graduating senior at the University and I'm going to tell you one fact about Julie and the fact is that she read fluently when she was four years old and I asked you to what is her GPA and the interesting thing is you know I can't see you because I'm blinded but but all of you have a number you have a GPA for Julie and as it happens this is a rare case with psychologists knows exactly what happened to you I can tell you where that number came from because this is a mechanism we understand and this is how it worked when I say Julie read fluently at age 4 you have an impression of how much that says about how precocious a reader she was you could you could do it in percentiles you know whether she fit in the distribution she is clearly not at the absolute top but she is clearly way above average in reading speed that impression is immediate and that's the way that we understand things we we have immediate standards you know for many traits when we're familiar with them that enable us to perform the struggle and automatically now how do we get the GPA from there well it turns out we also have a sense of the distribution of GPA and what people do is the GPA that came to your mind was pretty close equally extreme as your impression of our precocious reader Julie was you matched and this ability to match things across dimensions you know if you could ask in in New York or in another city how tall were the building how many stories would a building would have to have to be as tall as Judi was precocious in her reading and you can do that too yet we can match we can match across intensity dimensions naturally were very good at that this is this is part of the machinery of the associative machinery and it gets used in answering questions that were not asked so in fact what happened is when that GPA came to your mind you were not answering the right question you were answering the wrong question how precocious was a reading and you were matching across dimensions to a GPA that is as Extreme as your impression was by the way statistically this is completely wrong this is absolutely wrong it's a serious mistake it's about as serious and it's about the same kind of mistake as neglecting what we call the base rate in the case of Steve the librarian but that's the mechanism so in a whole set of situations we ask ourselves questions and we come up with the answer to a simple question it comes up automatically through an associative process it is not random we don't answer any old question there must be a connection between the two precociousness intelligence GPA they belong to the same context and we match we shouldn't it's statistically wrong but this is the way we do it let me conclude the mind that I've described and are really focused on system 1 and sort of assumed that in many cases of the system 2 endorses the suggestion of system 1 if you notice and all the examples in the last 10 minutes that were the case we asked a question that sounds like a slow thinking type of question but people's answers are really dominated by something that happened in their associative memory the basic analogy is there it's the visual system that is the way I think about it with guides my thinking about fast thinking is that a lot of the many of the rules that apply to visual perception extend in a fairly natural way to the domain of intuitive thinking to how we think associatively and intuitively that's I'm anticipating something that is my one of my reasons I mean I'm pessimist you know I'm a pessimist my temperament I introduced myself as a pessimist but one of the reasons I think that it's difficult to change the mind the way the mind works is that you have to imagine changing the way that our visual system works and that is certainly very very difficult and I think that changing the way the mind works may not be a lot easier so that's that's a story I wanted to tell you I'm just wondering I want to give you an example of how that can be applied because you know this I think that this kind of analysis lends itself to many applications and I'll give you I'll spend a few minutes answering a question I haven't been asked yet which is how does the two system mind know things what does it mean for a to system mind to know things and I actually anticipated the answer earlier to know something you know in the language that there's a proper definition of knowledge somebody said to know that X if X is true and the person believes that X is true with high confidence that's called knowing that's a definition of knowing the psychological state of knowing is different psychological state of knowing is when no alternative comes to mind it is when you have a single alternative that fits that is coherent with the context and it fits with everything else and no alternative then we feel we know whatever it is that we are thinking at the time in the same way that I thought I knew what my wife had said when in fact I didn't know it but subjectively no alternative came to mind that's what she had said in fact an alternative existed and the key here is not tricking it's not being aware that an alternative might exist and that is the case in the perceptual system and it turns out to be the case in intuitive thinking as well so that I think is what explains or contributes to the explanation of why people think they know that their religious faith is correct or their political position is correct or their beliefs about other things and they really have no evidence for and their their beliefs are correct it's because the intuitive system the societal system doesn't deliver an alternative and in order to modify that system so we had a conversation with Stuart just before I before the lecture we should have it earlier I would have prepared differently but it's clear that Stuart and Danny Hillis with with whom we were talking earlier they they believe that the mind can change they believe the mind will change and that if the mind itself will not change it can be so it can be augmented by by instruments that will help us think better and that's the way that one would think we know of examples where this kind of augmentation clearly works we know for example that pilots can be trained to fly by instruments that's a very different kind of flying than flying by the seat of their pants all right flying by visual perception is it possible to imagine augmenting people's cognition by the equivalent of flying by instruments fascinating question I don't have an answer I mean you know my my inclination is to be pessimistic but but i but in this case I think I'm wrong I think there will be ways in which we can improve thinking well augmenting it by possibly by suggesting alternatives so you could imagine in some situations that there is a device that whispers in your ears that your interpretation is not the only one now if we would be terribly uncomfortable with such a device then you know but there are occasions when you're making when you're in an important situation when it's very important not to make mistakes when mistakes are costly that's when somehow you would consistent do to be more active and that's where somehow we would need the equivalent of flying by instruments or at least instrument that would guide us in how to fly and how to think we're a long way from getting there thank you [Applause] I'll explain that that's weird got your attention in a different way there didn't we so much of your research seems like it's been done in collaboration obviously with Amos Tversky and with Richard thaller and others that you mentioned in your work say something about collaboration versus old research well I have mostly you know most of my work has been collaborative well I'd like to talk with people it comes back for me to me and I like doing this in in my collaboration with Amos Tversky which is you know the basis for the book really not that he would have liked the book in its I'm not sure he would but why what do you know well because we had that that was what I was about to say with that we were different so we were different and yet similar enough to understand each other but different enough to surprise each other and to present you know different points of view and we liked each other so much and we were so practiced at working together that we were enriched by those differences so we had different tastes but when there was something that both of us liked it was bound to be pretty good so there was - there were those two filters kind of a Venn diagram approached you for there's overlap here you're really onto something did that work I remember at one point you collaborated with a real sort of opponent in the whole psychological world know that I hate controversies you know that's there's a lot of emotion in science I mean it's not they're people who enjoy controversy that people hate it and I belong to the latter group and I hate being angry actually that's that's really what happens you roar when you get angry one occasion yeah but on that occasion there was somebody is someone it was become a friend whose position on issues of intuitive thinking is really about the hundred and eighty degrees of mind that is he his life project is to show that expert intuition is wonderful and you know I don't know how you'd characterize my life project but it certainly is not in that direction I mean I've been I've done more work on flaws of intuitive thinking then on the marvels of intuitive thinking but we collaborated in trying you know we both agreed intuitive thinking sometimes flawed and sometimes marvelous and we collaborated in trying to find the boundary and I think we did a truly I mean you know it wasn't revolutionary other people had had come to approximately the same conclusions but we did that collaboratively over a period of years that was not the same thing as you know as much fun as it was with Amos Tversky but but it was profoundly satisfying because we did we did manage to a green in fact the title wallpaper was I forget with the first part of the file is but the second part is a failure to disagree we there's a question that came in from a live stream from Molly know what does humor do to change or teach system one or his humor related to system two surprises interruption or stereotypes all of that and there's some lot of study on what makes a joke a joke I mean clearly there isn't you know I'm not an expert on that but there is clearly an element of surprise in humor that that is part of it humor is profoundly important to the workings of system one pleasure is fundamentally important to the working system one it turns out we think they're friendly when we're in a good mood or when we're in a bad mood when we're in a bad mood system two tends to be more active we're more vigilant we check ourselves more when we're in a good mood we're more intuitive we tend to be more creative so there's many associations between humor pleasure and the workings of intuitive system so the way to trick people in system two is to make some jokes that get them into it that's that I think by the way is one way when you want people to loosen up and come up with ideas you could do worse than telling you Joe sahel yes have you found any evidence that long-term use of stimulants such as Ritalin I hadn't all coffee we talked about beforehand or SSRIs effects or changes that balance between people's operating system water system - I don't know it's I just don't know the answer to that question okay and be fun to do the research yep with some of the substances we were saying to each other before Hanna's we were clutching her a cup of coffee that there's clearly something going on with the caffeine and you spoke kindly of nicotine I think that should be said right out loud yeah I mean I used to be a heavy smoker and I I don't think you know I I honestly think I was smarter when I was a smoke smoking but a whole people a lot of people's association just got jolted this is like this is like Hitler loving children and being a vegetarian Carter asks big wicked problems like climate change requires to understand the world and new and complex ways are we doomed because of the difficulty of overcoming our initial associative thinking now that's supposed to left of my heart if I think this analysis is really frightening when its implications for the reaction to climate change because my sense is that mobilizing people to action and in particular to cost the action involves emotion it involves system one you've got to speak to system one to mobilize people to act in significant way and the problem is whether you can arouse emotions and global warming as as a very you know frightening characteristic which is that by the time it's obvious that we're in trouble it may be too late to avert the trouble this is the kind of threat for which system one is really very poorly prepared so that I think there's then the sense of that question and I know there is a book david Runciman an important Englishman English scholar is is coming out with a book whether democracy can handle problems such as climate change and he raised us exactly I've been framing it can democracies handle climate change that's a century size problem of global size problem yeah and it's it it looks as if full dealing with climate change with authoritarian systems of traveler system might actually be better better able to recognize the distant threat and take action what we have in the United States it's really quite remarkable what is happening with respect to beliefs in nobody can really make a judgement about global change you know except the experts and the experts most of them almost all of them agree but where do I believe in the reality of global change its global climate really it's it's exactly an example of what I was talking to earlier I believe in the people who believe in climate change so you see other scientists say is probably right yeah and and you know there are other people who say that if somebody you know like a preacher they believe in says not to worry about that there are other worries then not to worry then there is another problem that arouses too that arises here which is we're trying to believe that if there are two opinions then you know who knows nobody can tell that very insidious that's you know there are there are two opinions there are people who are saying the other thing scientists may know that's you know there is a huge difference in both in number and in reputation between those who favor one position and those who favor another the public doesn't very difficult four people asked what happened after you ward I forgot to tell you they were very puzzled and and they stopped in their tracks and they walked back to their car that's a successful ELISA revoir CEO has a question that always comes up how do you think our brains increasing codependence that the Internet is impacting the uses of system 1 and system 2 if any what is you know what you see is all there is is both exacerbated in a way on the internet where you can see the beautiful thing the same thing but it's also there's a lot more leakage it's that's right I mean I don't think that this is a simple answer we are we're you know in a world where we we do have developed the habit of looking things up you know looking things up exposes you to two sources and I I think that's you know this is a profound change that have occurred I don't know how to describe it in terms of system 1 and system 2 but looking things up as as a way to resolve differences as a way to figure out what is really going on and then there are trusted sources then there is you know Wikipedia as only there is something we all trust and you look things up that's that's important you know that's going to have an impact on [Music] hope what's going on with our physiology right now this event was cut a little short by a fire alarm in the building [Music] [Music]

Channel: Long Now Foundation

Views: 31,069

Rating: 4.8730159 out of 5

Keywords: Science, Psychology, Culture, behavioral economics, prospect theory, intuition, fast thinking, slow thinking, nobel prize

Id: gmjgZF2HEwI

Length: 77min 37sec (4657 seconds)

Published: Fri May 01 2020