Nobel Conference 2008- Robin I. M. Dunbar

good afternoon ladies and gentlemen thank you very much thank you very much for the invitation to come here it's a rare privilege uh about a month ago i was at a meeting in london and i happened to mention to somebody that i was going to some rather obscure part of north america that i've never been to before to a conference there and he just went gustavus the nobels so you'll be glad to know that the fame of these meetings has spread across into europe the other side of the american channel okay i'm going to give you kind of a rather uh broad talk in a way and try and pull a lot of different things together the kind of story i'm going to try and weave for you and i'll try and show you some data on the way just to prove that it's not all speculation is really why we aren't just great eights we are great apes biologically that's our family and indeed we're buried right in the middle of it we're not even kind of on the edge we are our nearest biological species as it were the chimpanzees as i'm sure you well know uh and we're much more closely related to them for example than the two subspecies of gorilla are related to each other uh so we really are buried very deeply but clearly even though primates are incredibly smart and and you know we know uh you know here's candy for example the famous uh uh bonobo who can communicate with uh his handlers through his uh uh talking board as it were they can do all sorts of clever experiments nonetheless in the end uh as i'm sure you well know here they have not yet sent anybody to the moon and there's all sorts of other things clearly that we've done which they haven't and i'm going to just finger these two here this is literally fingering it isn't it uh religion and storytelling as something that demarcates humans in particular and they're kind of iconic of what i think is the big difference between us and all other species and this is not to set us on a pedestal and say we're special in any way in the end we're just another unique species but it is to try and understand why it is we're not just another great ape that in the course of our history and i think that comes down to our ability to live in the kind of virtual mental world of the imagination inside our own minds and it's that capacity which has given us that it is that capacity which has given us science it's our ability to step back from the world and say i wonder why it's like that could it be another way and that's what writing a novel is all about here's another way of the world could it be no other species of animal i am absolutely convinced can do that their noses are thrust right up against the grindstone of the world and they simply can't step back far enough to do it okay so where does all this start it starts with the social brain hypothesis which is an explanation for why primates have unusually large brains for body size compared to any other group of animals and the proposal which was put oh 20 years ago now is that because primates live in unusually complex social systems their societies are dynamic complex machiavellian was the term that was used originally although a lot of people haven't liked that and that kind of constant dynamic changing pattern of relationships or just requires a big computer to handle it so some of the evidence that was put forward to support that eventually is of this kind this kind here you can plot something like social group size as a crude measure of social complexity against some measure of brain size we've always used neocortex ratio or relative size of the neocortex because that's the bit of the brain that's exploded in size in primates uh your neocortex a it's a sheet about a a yard square it accounts for 80 percent of your total brain volume and the reason our brains are so convoluted is trying to wrap this huge sheet uh around the inner core of the mammalian brain and it's a it's where all the clever stuff goes on basically uh if you do that then you get these nice grade lines here where within grades of sociality is what it turns out uh the size of group you live in is a function of the size of your neocortex and down here at the apes who seem to have to work harder computationally if you like to support a group of a particular size than say any of the monkeys well that was the story we first proposed that and showed this evidence to support this in 94 two years ago uh cezanne uh actually now i'm jumping ahead of myself let me just show the kind of complexity that uh is involved or that people sort of uh viewed here and indeed you can look at lots of different behavioral measures of complexity for example here the amount of what's called tactical deception uh plotted against brain size again it goes up whether you use coalitions in your relationships allies to protect yourself species that do have a much bigger relative neocortex size relative brain size than primate species that don't and just to illustrate these kind of complexities i'll give you two examples here this one is a very well known one some of you may have read about it it's hans kummer's observation of hemodrasta brooms hemodiaspoons live in ethiopia they live in large groups family large groups of families and each family in effect is a male with several females and the males guard their monopoly of these females absolutely jealously um and kuma once worked what watched a female inch her way behind a rock here took her 20 minutes to get there and then when she was behind the rock she started grooming with this young male from the family next door now kuma says they weren't did this for quite a long time she never seemed to let her guard down is how he puts it and drop her head below the rock which she could easily have done she always seemed to position herself so her head was visible acro across the top of the rock and he his sort of charitable view of smart primates if you like was to say well look what she's really doing is saying so long as the old fool can see my head he'll think i'm just innocently sitting here these are some of my own study species the gelada this is family rivalry sibling rivalry really this is mum here she's got a head down she's being groomed by her younger daughter this is her older daughter now you remember this those of you with little sisters or big sisters just look where her hand is right she's got it right behind there she's trying to ease her little sister away so she can get into groom mom because being mon's grooming partner gives you advantages but look what it where her head is she's looking down very carefully watching what mum's response because mum is kind of asleep being groomed like this she's very relaxed and you know sort of half asleep watching very carefully to make sure that mum doesn't know because if mom realizes what's going on there's hell to pay is that kind of social complexity that's an issue okay so that that was the story until about two years ago and then suzanne schwartz and i tried to look at a number of other groups uh to see if of animals to see if this story held there too people have been trying to do that with very mixed results and we got a big shock because it turned out that in every other group we looked at and we looked at uh three uh big families of mammals the carnivores the deer antelope group the ungulates and the bats as well as a very large sample of birds spanning the whole spectrum of birds about 137 bird species in here covering almost all groups in every single case it's species impaired bonded monogamous relationships that have big big brains right so this is relative brain size here uh and the line is the average for the group if you like if you're above it you've got much bigger than average and these are essentially solitary species here uh monogamous pair bonded species kind of harm living species and species that live in very large multi-male multi-female flocks or herds and you can see that for all these groups it's pear bonded species that have big brains the only exception to this is in the primex and do what you like to these data you cannot get this quantitative relationship between group size and brain size to disappear and you don't really get a very strong signal of any kind of monogamous effect there's a weak one at best and this rather surprised us it was not at all clear what was going on but clearly at some point in the course of primate evolutionary history and that means very early on because remember the primates are one of the oldest unchanged largely unchanged groups of mammals in the world their ancestors are kicking around with the dinosaurs well actually everybody's ancestors are kicking around with the dinosaurs but these have changed least among uh certainly these other widely distributed groups of mammals what we know is now is that this has to do with social with bonding or pair bonding in monogamous relationships because the birds do that very nicely for us because in the birds you can split them up into those that have lifelong pair bonds those that have annual pair bonds like many of your garden birds which repair each spring and those that never never have permanent pair bonds it's the lifelong pair bonders have unusually big brains when you partial out anything else you want to get rid of and these are all control for those of you who are bothered about these things i'll just say these are controlled for phylogenetic relationships so the question is why are pair bonds so demanding cognitively what is it about a pair bond uh that makes it so demanding we think there are two suzanne thinks it's this stuff down here she's a girl uh and she's worried about you know sort of choosing the right mate and mate quality and it's clear that you know if you make a bomb choice particularly if it's a lifelong pair bond which you in effect can't get out of because it's kind of you're locked into it then that's not a very good uh uh idea doesn't do much good for your biological fitness i actually tend to think this is the real problem here is behavioral coordination and synchrony over the course of every day of your life because the problem with this stuff is once you've made your decision you picked a good mate you're away you don't have to worry about it again why should you need a big brain after that but this stuff this stuff this is an everyday problem and if you think of it in bird terms you've got a lot of very careful scheduling to do on a day by day basis when uh the eggs and the chicks are in in the nest because if you go down to the bar downtown and don't come back to four in the morning uh it's not such a problem for humans maybe but for birds that's a serious problem because your mate be the male or female cannot manage to survive that long without feeding so you've got to have a much more tightly coordinated scheduling you've kind of got to understand what it is your your mate needs really and factor that into how you organize your life so i really rather think that this is why this is a problem and with lifelong monogamous species that live together permanently this goes on day by day by day well you know that i don't have to tell you okay so where does this leave us with humans well we can plug humans into the same primate graph here the apes down here we can plug humans in for the benefit of the graduate students here i should just point out that this all this data on brain primate brains came from the same lab and as it happens in germany they had a few students that were late on their term papers so their penalty was to contribute their brains to science so we can put plug humans in at the appropriate neocortex size here and read across and it gives us a group size of 150 and this apparently is now known as dunbar's number which is slightly odd because i actually think it's an artifact i'll explain why do people live in those kind of groups i mean we live in these huge groups there's a great deal more than 150 of us here um uh and we can get on perfectly well in in these kind of large conurbations if you actually look at tribal societies then what pops out very clearly is that very often the number 150 is a core grouping size so these are census data from tribal societies all around the world including some american indian data from the last not the last century now the century before last the 1880s 1890s from california i think so each sort of column is is one tribal society and you can see they have a number of different kinds of grouping levels and these are nested within each other if you like think of it you know even now we live in these kind of social arrangements you know you have your family you have your extended kinship group that the family is part of extended kinship group's part of the parish the parish is part of a county and so on it all builds up and there's this one grouping level in the middle here which is absolutely perfectly pitched around the predicted value of 150 which is this red line the mean is very very close to that and these dotted lines here just the statistical confidence intervals around that prediction so most of the data ought to lie within that and you can see the red ones do they're they're of a particular kind associated very often the name like a clan or a regional grouping or something like that they're a virtual group you they almost never meet as a whole but they have very particular relationships with each other they're very different from these smaller groupings the band or these larger groupings the tribe and the megabands and there's lots of kind of casual examples people start to send me these things in actually after a while um i don't need any more we have a very long list of them uh but this is just a sort of selection of of other examples of this uh and i'll really just point out one well there's some nice american ones here's the mormons going off on their great trek and brigham young when he was sending them off i think there were five thousand all together that went off from the great trek to to salt lake city he he realized he was going to have a real coordination problem so apparently so i'm told and i'm told on authority this is actually what it says in in the mormon museum in salt lake city he split them into groups of 150 sent them off in little parties like that this is the one i really want to mention here the hutterites because they're just next door they're just over there in north dakota now remember the hutterites are communal christian fairly fundamentalist uh christian group that came here from middle europe with a very communal uh lifestyle so they farm communal communal farms everything is done communally obviously they all live in their families but all the work even the washing up and the cooking is often done communally this figure is just at the lower limits of the uh confidence limits around the prediction of 150. the reason it's 107 and and i have this simply from the the the hutterite data themselves citing the hutterites is that they will not allow their communities to get bigger than 150 and they would rather split them when they get to that number and set up a daughter farm somewhere else and start a new community and the reason they do that is very clear if you have more than 150 people in the community you've got to have a police force and their whole ethos is against having hierarchies and police forces and things like that so to avoid that problem they do it by keeping the numbers down that means everything can be done by peer pressure alone and people will respond to mild complaints about behavior for example for example by changing their behavior they won't say you know i mean in our communities if you go along and say you know we really didn't like what you did saturday night most people would say well you know what you can do with your complaint don't you if you've got less than 150 it's a very very different story because people have obligations to each other it's all a matter of trust and reciprocity you're bonded into this very intense you may fight like hell with them but still when the chips are down you will lay your life on the floor for them and that sense of obligation causes them to uh operate in this this sort of way exactly the same among the nebraska amish who paradoxically have nothing to do with nebraska they live in pennsylvania so this led us about looking to see if we could find real data on human social networks and this was our first attempt i rather like it because it's about christmas cards uh we just asked a bunch of people to write down the long list and some of them were very long of um names of the people who they were in the households they were sending um uh christmas cards to and you have a big range of variation uh here's tim robinson up here with his it's about 300 personal friends that he knows in great detail this is mr no mates down here that's me uh um but the me and it's very you know it's sort of a lot of variance in there but there's a very strong peak and the mean is almost exactly on 150 it's 153 as you can see and one of the other things we ask them to do is to say for every single person on that list when they last saw them how many months ago did they last see them personally and how do they feel about them on an intimacy scale from zero to ten so zero is i couldn't care less really and ten is you know they're my deepest something or other so this just plots time since last contact so this is yesterday this is several months ago uh and here we have this intimacy score from zero down here to ten up here and you'll see this lovely um correlation we've replicated this correlation on a number of occasions of completely different data sets and it's very very robust now the exception of these guys down here i just draw these to your attention because if you notice you see them very often but you can't bear them no they're not your family that's your family down there they're the people you work with okay if you look at these data a bit more closely you will notice that they kind of set off into a series almost like a series of subgroups here and that made us think that maybe actually this 150 that we all think about actually is is not in itself a unitary thing it actually consists of a series of sub-layers with different numbers of individuals in it so um i'm not going to show you the analyses we had to use fractal mathematics to to analyze these data in fact we analyzed two sets of data we got exactly the same story one was the christmas card data the other was our ethnographic grouping size data and although these days were completely different we ended up with exactly the same picture that your will consists here's your 150 consists of a series of oh we've lost the inner circles never mind there's this couple of inner circles in here and the size of those circles are very consistent so we're adding numbers as we go through there's you in the center this is your social world there's you in the center these are your five closest people some people have a little less some people have a little more but they're typically five and then there's another circle outside which consists of about another 10 people making 15 altogether and then you add some more in uh there's makes 50 150 and as you go down through this these circles the number increases the intimacy declines and the frequency of contact declines at the same time and we know there are two further circles at least beyond that there's another one at 500 there's another one beyond that of 1500. this is why i say which is dunbar's number because i don't know but the interesting thing is that they scale very tightly with this uh a ratio of of almost exactly three or a little over three which is we have no idea why it should be three your white should scale in that way but it's very robust somebody else has recently shown it with uh another set of hunter-gatherer data separate hunt together a data they got a little bit bigger figures than about 3.8 but it's in the very much the same ballpark and we've just shown you get exactly the same scaling ratio if you look at these hierarchically organized social systems of things like orcas the killer whales in elephants and again in many of the primates so it's clearly a widespread thing and it raises two questions really is what sets the limit on these layers in your personal friendship networks is it the frequency that you interact with them to have a very close relationship in here do you have to spend a lot of time with them uh and therefore the fact that the time in the world is limited unfortunately there's only a limited number of people you can afford to invest that amount of time getting to know if you like and continuing maintaining the relationship afterwards or is it a cognitive constraint is it that you only have five boxes at emotional intensity ten and if somebody else hose into a view in your neighborhood and you want to kind of bring them in here then somebody else has got to drop out and we have no idea you know what the dynamics of that is the problem we're currently working on just to highlight uh how widespread this phenomenon is here's this rule of three again in military structures all modern armies are organized basically on this principle everything's in sections of three you usually get some extra sort of headquarters bits and and other sort of accessories added on at each stage but essentially the main working parts you get three bits they add up and it all works very nicely now here's a test for you or i just point out the company right the company comes out absolutely spot on very close to the size of 150 just a little on average just a little over taking all the armies of the world which which this is based on um the army is the the company really is the fundamental structure it's the smallest unit that can stand on its own two feet independently and it seems uh the rest of it is sort of built up kind of as it were out of that i wonder if you know who's who really created this structure for modern armies this gentleman here you don't know who he is surely you do of course it's davis adolphus himself oh in his in his period during the 30 world years war in europe he was the finest military general and uh a military thinker uh of his time and and much admired afterwards and i'm told much admired as a as a for what he did back in in uh sweden as well uh as the king of uh sweden the hennessey's often regards one of the greatest kings of sweden which is why we have this uh place here and i the president has actually just asked me just to point out to any gusties here that if you didn't get that you're getting a fail grade all right how do primates bond their social groups well it seems to be a dual process one of it involves this stuff grooming now grooming is very intense for them they spend a huge amount of time doing it it's not just about removing bits of muck from fur it goes beyond that it's a bit like kind of cuddling and petting a massage in humans it actually has the same effect as as those it's very pleasurable when you're being groomed groomed by primates much as it is when you're being petted and cuddled and so on uh by by a human so that's this sort of emotionally intense component and i'll show you in a minute why it seems to have that effect but the other component is clearly this and what i think happens is it's just a guess really at this point is you need this kind of physiological mechanism to set up a kind of internal pharmacological state which allows you to build a cognitive relationship of trust and reciprocity somehow but we have no idea how that really works clearly however it has something to do with this stuff and john just showed you one of the classic tests the so-called salient test of false belief everybody assumes that this kind of special social cognition is what really makes uh the world tick for primates and what demarcates uh them in particular and we have it in spades and this uh former cathedral cognition is often put in the form of orders of intentionality intentionality is having beliefs about uh mind in your mind as well about uh other individuals or the world out there and so you can sort of structure them in in a sort of set of sequence reflexive sequence here's jack he believes that x is the case jill here believes that jack believes that x is the case so she's in second order she's got handling two belief states in her mind uh jack here again now is in third order because he believes that jill believes that jack believes that something is the case so he's handling three and this in principle it goes on beyond that uh uh uh essentially in an unlimited direction in principle this one here is very critical second order intentionality it's theory of mind children acquire that about the age of five uh that's the point at which children learn are able to understand other individuals have mind states which differ from their own so that's the point at which they learn to lie with conviction they also the point at which they can engage in in fictive play and pretend play that uh you know pulling a a a a piece of string behind you with a with a trainer car attached to it or whatever maybe it's the car really is going along they can have those kind of imaginative things so we don't get science below that come what may that's that's if you like the mineral because at that point you're standing back far enough from the world to imagine that the world could be different it set us wondering just what the limits are for normal adult humans so we ran a series of sort of extended uh false belief type tasks running up to ninth order i think one set ran up to ninth order this is uh a set we did up to sixth order uh with peter kinderman it's actually the first data set where we tried this so you have these little stories about people trying to make arrangements for dates or get a pay rise or something like that thinking through the mind states of the various characters involved and then we have a control here which is just factual questions about the story and the the general setting of the story that have no mind states in it and you can see that even when they're trying to causally relate a whole up to six or seven different facts in a kind of this kind of sequential uh situation they have no trouble at all um but when they're trying to handle mind states when you get up to fifth order is about the absolute upper limit it really plummets after that uh they tend to find these um uh mentalizing problems extremely difficult after uh fifth order so this is the average and we've repeated the situation lots of different ways we keep getting fifth order as the limit something like 70 of people have their limit uh at or below here and just to give you a sense of what fifth order is about here's a little example that i'll let you read okay if you got that you get an a grade if you didn't get it too bad okay so they're actually quite difficult to handle and there's some sense that maybe without language we can't actually get up to fifth order it'll actually be quite difficult to do but children go through this progression where they acquire second order about age five and then third order at about age nine or so fourth order at about 10 11 and they get up to adult levels in their early teens well we asked one group of people as well as doing these tests to also make a list of their inner circle of close friends and this is what the data look like so here's the number of close friends they had here's their scores on these mind reading tasks the limit that they could cope with the upper limit so some people didn't do very well first order uh uh some people claimed actually it went up to ninth order and i have to say i didn't write these uh i my limit was actually about sixth order i wrote the first set and it nearly broke my mind to do it frankly um but the graduate student that did this particular study actually claimed he could write write up tonight thought but in general forgetting the tale ends it doesn't really matter there's a nice relationship here the number of friends you have increases with your competencies on these kind of social cognition tasks there's a lot of variance in the data we know where some of that variance comes from some of its personality differences some of its age effects some of it's a gender effect i'm not telling you which one does better than the other that's because half of you know anyway and the other half couldn't care however what is really scary is this stuff shows up in your brain as you might expect it really does so this is some imaging studies we've just finished uh last year they're unpublished so you're getting the first really the first public uh airing of this stuff this is just showing that uh these mentalizing tasks as you go up through these orders it really is neurally taxing it's hard work for the brain to cope with them what we've shown is that in particular in these three areas here which are also pitch up as the three core areas associated with theory of mind competences uh second order intentionality i believe that you think that something is the case which is where most of the work has been done in fact all the work has been done when people are being tested for those kind of competencies you know what five-year-old kids can do for heaven's sakes these are the three areas that keep coming up for those of you who really want to know that's in the dorsal prefrontal cortex this is the temporal parietal junction just back here and this is in the temporal pole just up in there so and this is the right right dorsal prefrontal cortex as well whereas these two it's usually both sides but the key point here is that as you go up through the orders of intentionality between order two and orders uh order six that we took them up to these areas are working much harder at each level they're actually working harder and they're recruiting more neural material as they do it as well so the brain one of the reasons clearly that uh we need a big brain is this kind of mentalizing ability that allows us to think through the implications of social the social world we live in is very very hard work we also did a structural analysis on a another set of people and what comes out very nicely is actually the number of close friends you have in these inner circles correlates very nicely with the size of this area here the dorsal prefrontal cortex so now we don't even need to ask questions of people we can just go and measure this and we know all about you of course there's lots of individual differences in there and it's not just all phrenology which is what it's in danger of looking like okay so back to the grooming component now that's the sort of cognitive bit back to the grooming component if you look at time spent grooming in primates against social group size these are species averages here there's a general increase in time so given to grooming so as your group size increases you have to groom more you're not grooming more people actually you're grooming your close buddies much more as though to make the relationship work even better uh it's on a pretty much a linear trend and we've reanalyzed these newer data and we get exactly the same result but it kind of levels off here at about 20 percent of time so these are gelada out here they're huge groups uh it looks as though there's a kind of upper limit here that you can afford to spend in this otherwise useless activity when you could be feeding for example um this is clearly just an ecological effect and you have to feed in real life you have to go out there and work you can't spend all your life engaged in social interaction so this sets this up a limit here in primates in general just to show you what's going on in terms of grooming i mentioned it's very relaxing it makes you feel kind of very happy and so on it's because it's extremely good at releasing endorphins endorphins is a brain zone painkiller they're related to the morphine uh family but you they're non-addictive they're slightly different chemically and and you don't get addicted to them in the same way that you do to the artificial opiates and this is a little experiment done some years ago by barry cavern a colleague of mine in cambridge in which they showed that if you give very very low doses of morphine so small that you wouldn't be able to detect them in blood tests or immediately afterwards interesting grooming plummets compared to here's a saline control they're just given uh salt water basically uh compared to these you're interested in engaged in in social interaction so the amount of grooming or the amount of number of grooming partners you have or the amount of grooming you ask for drops in contrast if you give them an opiate blocker in this case a a fairly inert drug called naltrexone which simply locks onto the opioid receptor sites in the brain and won't let the endorphins get in there you just can't get enough you go crazy because you're not getting the the release from it from the grooming that you would normally be getting so you just keep asking for more and more and more and this may be one reason why uh morphine drug addicts become totally asocial because they're just it's not spaced out they're blocked out they don't need social interaction to get this buzz that seems to underpin our relationships okay so what does this the effect of this for humans well if we plug humans with our groups of 150 into this primate line and we we ignore the fact that there's an upper limit uh we would have spent about 43 of our day time engaged in social grooming if we were to bond our groups the way primates do so this is a big gap here it's a lot missing because if you look at real human data and these these are times spent in social interaction by a whole range of people from people in in dundee and scotland to herders and maasai herders in tanzania hill farmers in nepal the average is absolutely bang on the 20 this is just variance around here remember these are averages for individual species so there's a lot of variants around here too these are bang on there's this big gap missing here the so-called bonding gap which clearly we don't have the time to do we're as constrained as monkeys look like we're using the same amount of time somehow using it more efficiently and you all know you know if you spend that amount of social time uh in your life nearly half your day engaged in socializing you would get nothing done i mean the only people that have that amount of spare time is academics let's face it so somehow we have to cut through the glass ceiling here and here's one suggestion that's what language was triggered to solve the problem of and language is very good because it allows us to cut through many of the blockages that limit grooming you can exchange information about the social network monkeys and apes know only what they see whereas you can go around telling people what's happened and catch up with the news you can talk to more people at the same time so here's a standard conversation group uh we can talk to four or five people uh at the same time a little conversation group of four or five people works very well it's actually the upper limit can't get more people into a conversation than about that but still that's considerably more than grooming for monkeys which is a very much a one-on-one activity we can talk and walk we can talk and do other things we can use it to share knowledge that's creating a sense of community and bonding but there is a problem with language however wonderful it is it's very poor at expressing emotional content as you know it's probably the reason we admire poets so much is that they can put in words stuff that about our inner feelings that we mostly really can't do we're very poor at expressing the emotional content but it's also worse than that it doesn't arouse any emotions most of the time it doesn't solve the endorphin gap particularly but at least not particularly well so what might have solved that well here's three suggestions as to what's happened on the way and this is more by way of hypothesis than anything else so what we have here is uh time into the past here we are now uh this is predicted grooming side time for all fossil hominid populations um here's lucy and herlock down here we're up here somewhere where this red line is um and it's simply based on on taking the fossil cranial calculating brain size from that and then interpolating these values through the graphs into neocortex size from the aquatic size to group size group size to grooming time and you can see here's the twenty percent the australopithecines they're just they're you know in the same border that's fine and then it starts to take off we're probably not too worried sort of at least down in this sort of sector here we can have a little spill over but certainly once we get up here the required social bonding time is far beyond what we can afford to do so we need something really to to to cut through this gap before we get to language and likely language only came in sort of somewhere around here perhaps at the latest uh half a million years ago with the appearance of archaic humans i think these three are fill in the gap they're all unique to humans with the exception of this which we share with chimpanzees laughter they have a form of laughter they don't use it as intensively as we do but laughter music and i mean the performance of music not you know sort of constant music in the sense of sitting well the orchestra on the stage is doing what we're interested in as it were but the the audience really isn't i'm talking about engaged in performance of music as in traditional societies where you might sit around the campfire and have a sing-song and also religion and particularly the rituals of religion not necessarily the intellectual content the the um theology as it were so and i think they probably came in in this order that laughter came in first simply because we shared that with chimpanzees so it's already there what's interesting is about laughter is it's also highly synchronized and segmented um and dancing and music performance as orchestral players will know is very dependent on the ability to synchronize and segment your behavior as it were because that's how the music is structured and so is language so i think we go from here to here and a naturally a very quick jump into language once you've got this kind of vocal singing as it were without words uh in the in the context of of music and dancing it's a very very short step in into to language so i just want to also point out and i'll just show you a bit of evidence in a second that all three of these are very good at triggering endorphin release and i'll just show you evidence uh for the sake of time for here this is the little it's actually two studies pulled together that we did a couple of years ago looking at the impact of laughing on pain tolerance so your pain tolerance is simply a good assay for endorphin release because that's part of the pain control mechanism and all we did was we showed people uh we first of all gave them a pain test uh sometimes we use a an old-fashioned blood monitor a blood pressure monitor sometimes we use a hand cold hand in a hand in an ice bucket anything like that will do see how long they can stand this show them something like a video or get them to do something and then test for their paint tolerance again if you show them a very boring boring tourist video happen to be from nevada actually not nevada nebraska but i take the blame for that because my granny was born in fontanelle so um there's no difference i mean there's a lot of variants around here but um uh you know the average is absolutely smack on zero there's no real effect if you do it with a comedy video and people are in groups to laugh they're all in groups of four and you measure the amount of time they spend laughing then even without that point down there there's a very significant relationship and we've shown this now with drumming circles we've shown it with singing in in church services uh and we've shown it in rowing eights uh you know these eights on the river and what is interesting about that and particularly interesting in the context of creating what i think this is doing is creating this upsurge of a sense of belonging to a community what we've been able to show both of the drumming data from drumming circles and with the rowers that if you do it on your own you don't get such a strong effect if you do it as a group and particularly a synchronized group as in drumming or as in rowing where everybody has to be absolutely in time together then you get this huge upsurge and you come out of it with this sense of real belonging to this to this little community and it has knock-on consequences and again this is an unpublished study although it's it's we're just trying to replicate one bit of it before we we send it off for publication if you do that same kind of study and then you put the people into a public goods game where you give them uh say in pairs you give one member of the pair five dollars or something and they can offer the other member of the pair any proportion they like between zero and 100 but the other person can accept or decline it and if they decline it you neither of you get anything what you find is that people who watch a neutral video here or a comedy video that come with a friend uh as their pair mate show no difference at all but if you come with a if you come in a paired up with a complete stranger then watching a comedy video and laughing together massively improves your willingness to offer in fact you you give them 50 50. if you watch uh you know boring video uh then you well the average is something below a one to four uh payoff ratio split you're much meaner to them be happy that's the answer okay so the question is why don't you go out and do it on its own i mean if it's just endorphins you know go up you know lots of people do just around the corner here you can go and watch them on the way to coffee the answer is i think if you do it on its own you don't get this huge upsurge that you get now the question is how do you get people to do that and i think that's where the theology part of religion if you like comes in that it makes you keep turning up as a group on a regular basis to engage in this kind of communal bonding process and i think it comes this is where the theory of mind components become very very important because if you do a kind of uh task analysis of what kind of religion you get out if you have only third order or fourth order or fifth order intentionality you get a very very different kind of religion the third order it is purely personal it's your beliefs and pretty much that's it at fourth order it's pretty much you can persuade somebody else about your beliefs but you really need a police force to make them go along with you at fifth order you're automatically sucked into both agreeing and in a way which you can't really get out of very easily it's kind of hard to see this i'll illustrate it with storytelling the other half of where i started because it's much much easier to say it's absolutely glaringly obvious so here's shakespeare's story othello you all know this story uh ziago uh telling othello these wicked lies about desdemona down here what are the cognitive uh processes that are going on here in terms of mind reading well if you think about it the audience believes that the argo intends othello to suppose that desdemona is in love with somebody else okay so at that point the audience is operating in fourth order intentionality they're having to cope with four sets of mind states on on the stage at the same time but this is not a very interesting story this is a jeffrey archer novel you know it's kind of eternal triangle but it's not really you know it's it's a quick read and it's fun and so on it's not very demanding and anyway who cares if why should othello care if desdemona is in love with somebody it's just a fantasy you know everybody has fantasies it doesn't matter what makes the story bite is that iago persuades othello that cassio reciprocates desdemona's love that they they're going to run off together and at that point you get beyond a geoffrey archer novel into a shakespeare play but the audience is now having to work at fifth order they're working at their limit and if you think about what that means is when shakespeare sat down to write that play he was having to work one level higher still so he was having to work at sixth order and now we're talking about special people so this is my pitch on why we can all enjoy stories and the more we're pushed to our limits by the author and getting us up to our cognitive limits the more we enjoy the story the more we get out of it as it were the more it drags us in but to be able to do that you really have to go beyond the levels that most normal people can cope with and that's why you should value good authors just so just to finish off very quickly let me go back to where we started and show you why that's only possible for humans this is very sketchy data but it's all we have i'm afraid if you look at these competencies and intentionality we know humans are at fifth order we know all the monkeys and everybody else seems to be at first order everybody's pretty much agreed about that the eights and the black daughters chimpanzees the only ones we have dated for just about cope with second order they're not very good they're probably about as good as children at age four whoa they're about as good as children aged four uh they're not as good as children who have full um here's my rescue full theory of mind competence there we go thank you very much guys they said if i did that it would turn the whole thing off they didn't realize i was going to test to see if they could do it good okay what is extraordinary and these are the uh open dots so simply the brain sizes or frontal lobe volumes of all the other species we have of monkeys next we have data for and you'll just notice they cluster very very tightly around these two points but these three points here just lie on a very straight line and if that's the case this big difference here is huge and if you look at actually frontal lobe volumes in effect in humans so this is total brain volume for primates split into uh on the bottom here the visual areas which are at the back which are a huge chunk of your brain and the sort of bits more frontal to that it's a bit more than just the frontal lobe here but we can't split it any other way what's interesting is this huge chunk you use for vision increases a bit with brain size but not a lot but the rest of the stuff up front and that's really the business end for this kind of cognition is just going up through the roof and so it may be no surprise that it's only when you get to the brain size of eights that you start to see these early stages of higher orders of intentionality second order intentionality and it's no surprise surely that it's only humans that have the capacity to get up to fifth order and thus to have science the arts drama and so on and in the end religion too thank you very much if we could ask our panelists to come up once again questions for dr dunbar please write them down pass them to the center aisle the ushers will pick them up we'll resume here in just a few minutes here 24 okay oh okay ladies and gentlemen would you please take your seats i guess once again if you have some questions pass them quickly to the audience to the ushers otherwise we'll we'll begin by giving our panelists an opportunity to ask questions of dr dunbar where where do we start here dr pablo okay so i found it fascinating that that part of the brain as do you say dorsal prefrontal cortex correlate with the number of friends you have or your social group can one go to fossil forms of humans and ask for habilis or homorectus or neanderthals what were there numbers of friends i was hoping curtis was going to do that i'll leave that to you the answer is no because that's individual differences between um within species and you you really don't get enough you can't measure that kind of individual variation on brain size or brain bit size in fossils is the problem i don't think anyway you can kind of it's so imprecise anyway and then you can't measure that bit of the brain anyway even from the impressions in the inside it's not good enough so the best you can do is what we have done with the general group size analysis of predicting group sizes for populations as a whole where we can do it using the general equations by interpolation and that just gives you a ballpark figure but i don't think you'd probably get unluckily dr maria yeah just a a point and a question um in and i was interested in what you were talking about with the impact of grooming on endorphin stimulation um and of course in traditional societies they lack mirror technology so the application of cosmetics is a social activity unlike in the daily activity of people in our society where it's a private activity in front of a mirror maybe the analogy might be the local hair stylist um functions as a social activity of cosmetic production and stimulation of the head and the and so on but in traditional societies uh the application of body paints and so on is a social activity that in many ways mimics grooming because of bodily contact and it happens both between men and men and women and women but not across the sexes um have you looked at that activity to see if it has a similar reaction as grooming does in primates and i'm wondering if that could be a transition out of grooming for humans we haven't looked at it and it would be a great project for a phd student um but we know that i mean some of our data come from hair care networks as we call them in in the kung sam some of our later on on the inner cliques and those and in those kind of societies where you don't have a hairdresser group small groups and i guess it goes on here in sort of more rural communities and you know even in in the developed western rural communities you have these little groups of friends who will do each other's hair but i guess i mean you know i think that's why when you go to the hairdresser you know particularly women where it's a long a longer process we just go in quick you know think run with the razor and itself but for women it's a longer thing i don't know ladies do you do you get this kind of uplift and from the endorphins let me just come in with a with a question here from one of our internet viewers here to follow up on that how does the development of such social networking technologies as facebook or myspace impact human capacities for relationship he says i have over a thousand closed friends on my facebook side um thank you for contributing to our database because we're actually doing a couple of projects on facebook and myspace at the moment the mobile phone companies have got really interested in in these kind of social networks because it affects well it affects how big the space you really need to have just on things like myspace and so on the networking and it actually affects how much how big the technology has to be on these virtual world gaming environments it turns out they have exactly these kind of structures that we we found in ordinary people even on these virtual environments um yes people claim to have huge net and the other reason is of course the technologists think that they can cut through some of these glass ceilings by increasing the broadcast pattern that you can have because you can send an email out to you know tens of thousands of people literally that saves you a lot of time the real question is what's the quality of relationship you have with them because we are embedded in a in this extensive series of circles of acquaintanceship which goes on forever it stops with the six billionths and two people that are now alive on the earth and probably includes your pet as well on the way um you know it's the number 150 is simply the number of people you know as persons that if you bumped into them in hong kong airport at 3am on a stopover at the bar you wouldn't have even think about twice about going up to them and saying hello and how are you and stuff because you know who they are and they know who you are you know there's lots of people you know beyond that the the limit about 1500 uh is uh quite well documented now is the number of people whose faces you can put names to and that's much bigger clearly it's an order of magnitude beyond 150. so you can have large numbers of people but the question i have to ask you is what is the quality of your relationship with them and you assume that they're all equal but in fact not really it turns out not really so and certainly from all the works being done on facebook most people have um uh circles of 150 and that's about it dr van hoystein yeah thank you so much uh this was such a fascinating lecture and your answer now leads into a question that i wanted to ask exactly about what you call the quality of close relationship you may just think you talked about circles of acquaintanceship and then asked what sets limits to those is it you know time investment or is it some cognitive restraint so i'm just wondering is there also a third option or is the first one it sounds like the more time you have the more french you might have that are very close to you when i would say how do you how do you account for personal choices to have say very few close relationships that aren't very intense and the same kind of question towards the end when you talk about religions and various strata of community the bigger the community more you are sucked in but what about the quality of religious experience if you if you choose for more personal investment and and individual uh let okay take that last point first as i probably should have commented on that it seems to me that these i'm thinking here in terms of the evolution of religion and how it kind of got going and where it came from and it seems to me that it has very much that feel of being a small community thing for bonding these very very small scale communities and i wonder whether this explains why most religions most of the big world religions are so prone to fragmentation in terms of cultures and sex and that things like the house church movement that's grown up in protestant christianity in particular is kind of reflecting the same thing that these very big uh churches as well congregations are just too big for you to handle it's it's a much more intimate kind of experience that you're getting or that you you'll want and what a lot of these mechanisms are designed to kind of create this sense of belonging um and now i've forgotten the first question where you started well the distinction you might be a close circle of acquaintances between uh time constraints or time investment and time and cognitive constraints and how the balance of that works and how personal choices may affect that circle of close acquaintances the answer is we don't know i mean that is actually what we're trying to figure out right now with a whole series of kind of studies of social networks at this kind of level um one thing i i might sort of answer with a with a kind of uh caricature as well story and that is it seemed to me that and it goes back to the question about uh electronic media as well that you know with all these big multinational companies uh scattered around the world you know some accountants somewhere would figure out it's much cheaper to give everybody a um a video conferencing system and then they'd save on these very expensive uh business trips all around the world and a huge carbon footprint and this would be the saving grace now i bet a company that did that would be bust in a year because what's important when these guys go on business trips is not that they go and negotiate something it's the same unit this adage all business is done on the golf course it's not that they're doing business on the golf course they're getting to know each other they're getting to trust each other as individuals that's all and if they don't have that sense of quality of relationship when one of them rings up from let's say boise idaho hewlett-packard's big lab over this side of the the pond and rings up their colleagues over at the bristol lab in england and said listen guys we've got a problem can you help me out if they haven't gone you know got to know each other the response is going to be listen guys we got problems over here too you know sort yourselves out first whereas if they've got really to know and trust each other then they'll come back and say okay well we'll do our best it's not ideal right now dr stanford yeah i thought that was a really interesting uh discussion and uh the the thing that i would be real curious about is that figure of 150 for the group size which appears to be optimal in hunting gathering groups as that group grows bigger then you see fission and i think maybe if we could plot that on kind of on a map like we've been doing with the clovis artifact distribution we can see how that how that fission would work and why absolutely and actually this raises a big issue vis-a-vis the archaeological record both you know your end the recent archaeology and you're in the deep archaeology if you like because we can make predictions about if you like about what any given population species subspecies of ancestral hominids uh group sizes might have been but when you look in the archaeological record what you see is the overnight camp which is typically in modern hunter characters in the 30 to 50 range it's that 50 in a 50 grouping rather than 150 which is much more diffused yet that 150 socially is probably much more important than the temporary campsite now our problem is how do we get together from our two ends oh we need to go out more yeah absolutely here's a question from the audience uh do you believe language and religion occurred simultaneously i no and yes because it depends on what you mean by religion right um the the the view if you like it's no more than idle speculation uh from is that religion kind of is developed through a series of stages i don't think this is terribly controversial in the kind of religious studies areas what everybody this is sort of a consensus view that you started off with these kind of uh hunter-gatherer type shamanistic type religions and maybe you can talk about them when after t and gradually these developed or evolved historically and into the world religions that we know and love as it were that we have today um so my my suspicion is that these kind of traditional society shamanistic type religions were triggered it's just a speculation triggered by the fact that humans find it very easy to trigger these trance states we can do it by any number of different uh uh things you know um you know that's bill clinton's problem he never inhaled there are all sorts of ways you can you can do it dancing is very good that's why the trance dancing circles of the kung san produce these translates you go off into this world of the mind and you experience all these extraordinary things in there and it's it's you have to have an expert it's very very real as indeed all these kind of religious uh ecstatic experiences are you have to have an explanation for it so you you you then start to think in terms of if you like a theology for that that kind of explain it but that early stage has my guess is it came in possibly even pre-language but i suspect not very long before language came as a mechanism of social bonding that once you've got language however that changes the picture because now you can add a kind of theology of some sort to it which explains why it all works and that's part of the bonding process itself it also allows you to say listen you guys you've got to keep coming every week or whenever it is that you can make the arrangements which is what the kung sang do they use it for for repairing damaged social relationships in the community people get fractious with each other in the community and then after a while somebody will say let's have a trance dance on saturday you know so they all get together in the barn and they have the child's dance and that kind of repairs relationships in the community and then it sort of for a while and things deteriorate again um so that that sort of sense of very early religion i can see coming in without with without without language but i think to make religion work fully uh engage the community then it made a big difference to have language because you could explain why you have to do this kind of stuff okay here's a here's a question i think this must be from a teacher any way to build your ability to handle higher levels of intentionality are you paying money for this shakespeare um oh it's too late but but it's clear that individuals can learn we did actually did a project a couple of years ago finished it's not yet been published neither in which we tried to teach improve people's performance on certain kinds of cognitive components of theory of mind partitioned up into its component and see if we could train them to do better and by and large younger kids were the ones that showed the big improvements it's well known that the speed at which you acquire theory of mind how well you do as children are very much very young children depends on the number of siblings you have it's social exposure really so now i'm sorry it's too late question about could you describe experiments that could be used to test for theory of mind and animals what degree do other animals have oh this is very difficult we tried it with dolphins and uh this is a very illuminating story so my apologies if i tell it in its full because it illustrates how difficult it is to do we've done uh colleagues have done no we had done some with chimpanzees and colleagues have done some with chimpanzees and they're reasonably successful so somebody came along who was interested in dolphins a bit of a dolphin nut i may use the term um very keen to show that dolphins are very smart and so we sent him off to south africa with this design design is very simple it's based on the salient type task and it involved having two boxes which were put on the side of the pool one of them was baited by an experimenter and an observer was looking over the shoulders behind a little screen uh the um experiment observer went out of the room the boxes were moved out to the side the screen had been taken away boxes moved out to the side and the experimenter either swapped the boxes over or didn't right so if he didn't swap them over it's a true belief task was the observer knows which box is baited if they were swapped over it's a false belief to ask because the observer thinks the box over here was baited but actually now it's been moved over this side so the experimenter comes back in we get the dolphin to come right in front in the middle and the observer points to the box that she in this case thought was the one that was baited well the dolphins were phenomenal chimpanzees take about 80 trials incredibly slow to learn the basic tasks before you get to the test dolphins got it in eight right okay these are performing dolphins they got it in eight and they were absolutely spot-on perfect a great students every time didn't matter false belief true belief task mixed up they got it every right every time this is phenomenal nobel prize is loomed unfortunately i made the mistake of looking at the video but actually what made me think about it was there was one task with the where the observer made a mistake and forgot what kind of trial she was on and she thought she was on a false belief trial and the dolphin knew she wasn't right so this made us so we looked at what the dolphins were doing and they're so smart you know no skin off their nose intellectually because they it just shows how smart they were they were picking up on how the experimenter was dipping her shoulder if she put the bait into the right hand box she tended just to drop the shoulder refraction and if she did it with the left hand box she'd drop this shoulder so we had a group of naive people look at these tapes and they were nearly as good as the dolphins not quite i mean the dolphins got it every single time they got it right so we then uh a bit frustrated we decided we do with a new set of observers new dolphins new design much tighter control or the other thing the dolphins are doing was they would come up to the center point the observer would tap the box to show them which one and the dolphin would swing to one side and then drift down and of course the experimenter then was left in the difficult position saying did they need this box or this box and they were very nice people getting the benefit of the doubt anyway we tried it in in in florida with some some the dolphins that come in from the sea to one of these facilities petting and patting facilities as it were so they're free free-ranging dolphins and we're much much tighter constraints on these kind of what are called clever hands problems it's like comparative psychologists live in terror of clever hands problems which at least no matter how smart your experiments animals can outwit you um and they just couldn't do it so it's those kind of designs but the real problem is and the problem everybody has had who's been trying to do these kind of studies on um uh great apes in particular and and at leicester where santa barbara is uh as a sort of one of the meccas for that kind of stuff it is really really difficult to design experiments which allow you to be sure that the animal has actually understood what the problem is and is demonstrating whether or not they have the capacity to do it and it's not just that they're bored or disinterested or they just haven't figured out what it is you're trying to do and we have the same problem obviously with very very young children who who pre-pre-speech children again you know it's hard to design experiments because with adults you can explain the the basis okay i believe i read somewhere that the consumption of caffeine induces in increases endorphin levels and perhaps it's about time that we should break into small groups and engage in group endorphin level raising here for a few minutes thank you once again dr dunbar we'll reassemble at three o'clock

Channel: Gustavus Adolphus College

Views: 6,844

Keywords: Nobel, Conference, Gustavus, Adolphus, College, Saint, Peter, MN, Dunbar

Id: i98XpBFWPrI

Length: 79min 58sec (4798 seconds)

Published: Thu Sep 17 2009