The Infinite Monkey Cage - Maths of Love and Sex

Episode Date: February 8, 2016

Robin Ince and Brian Cox get romantic (although unfortunately not with each other) as they discuss the mathematics of love and the statistics of sex. They are joined on stage by comedian and former maths student Paul Foot, mathematician Hannah Fry and statistician Professor Sir David Speigelhalter, as they discover whether a knowledge of numbers can help you in the affairs of the heart? Can a maths algorithm help you find your perfect mate at a party and what do the statistics tell us about what happens after the party, if you do! They find out whether mathematicians are more successful at dating than comedians, and whether a rational, scientific approach to love and life long happiness is really the answer.

In our new podcast, Nature Answers, rural stories from a changing planet, we are traveling with you to Uganda and Ghana to meet the people on the front lines of climate change. We will share stories of how they are thriving using lessons learned from nature. And good news, it is working. Learn more by listening to Nature Answers wherever you get your podcasts. This is the first radio ad you can smell. The new Cinnabon pull apart only at Wendy's. It's ooey gooey and just five bucks with a small coffee all day long. Taxes extra at participating Wendy's until May 5th. Terms and conditions apply.

Hello, I'm Robin Ince. And I'm Brian Cox. And in a moment you're going to be hearing me saying, hello, I'm Robin Ince. And I'm Brian Cox. Because this is the longer version of the Infinite Monkey Cage. This is the podcast version which is normally somewhere between 12 and 17 minutes longer than that that is broadcast on Radio 4. It's got all the bits that we couldn't fit in

with Brian over-explaining ideas of physics. I do object to the use of the word longer, though, because that's obviously a frame-specific statement. Yeah, we haven't got time to deal with that because even in the longer version, we can't have a longer intro. Just let them listen! I've got an idea!

Can we just have a podcast version of this intro to the podcast, which can be longer than the intro to the podcast? Yeah, it will be available very soon. Hopefully it's started by now, but if you're still hearing this, I don't know what's going on. And then we can have a podcast, podcast, podcast version of the podcast and then it would be a podcast version. Hello, I'm Robin Ince. And I'm Brian Cox. And we're well into Series 13, as usual.

We're getting lots of lovely emails and questions about the nature of the universe, the definition of reality, the standard model, epigenetics and whether Brian uses shampoo and conditioner or an all-in-one formula. But the most common question that we always get is... I've been listening to Monkey Cage for 13 series,

and though it's greatly enhanced my ability to talk about weakly interacting particles in the Higgs boson, I still can't find love. Can you help me? Well, yes, we can, because today we're going to give you the power to find your part of choice using statistical analysis and mathematical formulas.

No more shilly-shallying with psychobabble books on men from Mars and women from Venus. We're going to get Euclidean in the boudoir. That's a meaningless statement, isn't it? Because how could you be non-Euclidean in the boudoir? I like to experiment in the bedroom. By whichoudoir? I like to experiment in the bedroom. By which I actually mean I like to experiment in the bedroom. I'm currently researching

blackbody radiation and I've got a very small photon gun as well. My wife's not happy. The curtains are both on fire and not on fire. So, this is a concept monkey cage. We'd like to invite you to a party. A party So, this is a concept, Monkey Cage.

We'd like to invite you to a party, a party just populated by mathematicians, statisticians and comedians. What a party! You may or may not meet someone, you may or may not go home with them, you may or may not get married, have a baby boy or girl, have an affair, get divorced and de-invent yourself as a post-modernist philosopher. We'll analyse the probability of each of these events and show you how to optimise your behaviour algorithmically

in order to find the perfect mate and live the perfect, happy life using mathematical logic and statistics alone. Let the party begin! So, swilling from the bottle of reason, dipping suggestively into the guacamole of appropriate doubt, our party guests tonight are... Dr Hannah Fry, I'm a mathematician from UCL,

and the thing I think is most attractive about mathematics is its absolute truths. I'm David Spiegelhalter, I'm the Winton Professor for the Public Understanding of Risk from Cambridge. And the thing I think is best about statistics is that you can strip off those layers of messy noise and reveal that elegant, smooth signal underneath. Oh.

Well done, David. These people are hot under the collar already. My name is Paul Foote. I'm a comedian. And the most beautiful thing about mathematics is a cute uncle. Thank you very much. This is our panel. Thank you. Hannah, we'll start with you, we'll start with the mathematics. How do you go about applying mathematics to something as,

well, I suppose what people would see as seemingly random as finding the perfect partner? Well, you certainly can't account for that sort of mysterious side of things. You can't write an equation for exactly who you'll find attractive. But there are all sorts of patterns in the way that you search for people in the way in the things that you or rather uh in when you decide to settle down in your life uh in how to optimize your searching strategies those kind of things and that's where the mathematics can really come in so can you give us an example of something that someone would you know there are people here mathematically inclined they're thinking i

haven't yet used it as a bridge towards, you know, love and future security. So the starting point. Give them a tip, is that what you're saying? OK, well, so, for example, there is a lovely piece of mathematics called the stable marriage problem. And essentially, you're at a party, and you have to imagine that there is a group of boys and girls

who are trying to target each other. I have to say a lot of the mathematics when it comes to dating tends to be framed in a traditional way just because it's a lot easier to have boys and girls or two groups of people targeting each other. So each person at this party has an ordered list in their head of who they'd most like to date. And if you allow it to play out in a very boy-meets-girl way,

you can follow that through with a mathematical proof and show that every single time, every person will end up finding a partner. But you can also prove that if the boys are the ones who do the approaching, they will always, always, always end up much better off than the girls will. And the thing is, that sort of goes against what a lot of people's strategy is when they're at a party, because to risk humiliating rejection by going up to people and, you know, seeing if they like you, it doesn't seem like a particularly comfortable thing to do. But the thing is, by doing that, what you're doing

is you're starting at the top of your list and you're working your way down. Whereas if you do the opposite, if you allow people to come to you, then you essentially end up with the least bad person who will approach you. So that, I think, is my first tip, then, is to be proactive. Math says be proactive. So to define by better off,

you mean that essentially we're forming an ordered list of attractiveness, essentially. Exactly. So 1, 2, 3, 4, 5, six, seven, eight, nine, ten. And you're saying that by being proactive and approaching number one and then approaching number two, you'll get higher up your own list. Exactly.

So even if you approach number one and they turn you down and number two turns you down and so on and so on and so on, because you're starting at the top of your list and working your way down, you'll end up with the very best person on that list who would even consider you as a smooch for the math. What's the branch of mathematics that you're using there? Because it's easy to see that when you describe it in words,

but you say that you can perform an analysis. So what actually is the analysis that you do? Because you can structure that as a mathematical situation, you can structure it using equations and so on, you can prove that these things are the case. It's following something called the Gale-Shapley algorithm, that process of boy approaches girl, girl decides whether she likes boy and rejects him if she doesn't

and accepts him if she does. And if you follow that through using these sort of equations, you can prove that these things are always true. But how do you know which ones are up for it? And also, wouldn't you be better off sleeping with all of them seeing which is the best one

and then choosing? That is another strategy, yes. The thing is, with this example especially, because it starts off in quite an abstract world, these equations where you have boys and girls who are approaching each other and no other sort of connections are possible, you have to be slightly careful about how much you pull that out of the abstract world

and apply it to the real world. But I think that the general message is quite clear and quite obvious when you put it back into words. It does make good intuitive sense that being proactive is more likely to get you what you want than standing back and allowing your suitors to sort of queue up for you. So what mathematics has discovered is that sitting back and doing nothing

leads to less results than doing something. Brilliant. Science is another monkey cage revelation. David, the subtitle of your book is What Statistics Can Tell Us About Sexual Behaviour. So, what can statistics tell us

about sexual behaviour? Well, a lot. But you do have to ask people. You can't put cameras in rooms because that might affect their behaviour, rather. It's probably illegal as well. It probably would be, but you have to go and ask people and therefore you have to sort of trust what they say. This is one of the problems,

that if you just put out a survey online or a magazine does it, well, people might reply. I don't know, people in the audience might... Who would fill in an online sex questionnaire if it was put up online? We want to know about your sex life. Well, no-one's put their hand up. No-one would be the furthest. They're not going to volunteer.

Exactly. Tell you what, if this was Jeremy Vine's audience... Oh, I bet they'd be like that. So if you want to get a representative sample and use proper survey methods, you've got to work really hard. The last sex survey done in this country cost £7 million to find out what people were getting up to, how often, with how many people. And this is useful information, believe it or not.

But why? Well, it grew out of the... It all started in the 1980s, when it started the AIDS epidemic. There was a panic, really, and people realised that they didn't know just what was going on. Who was having sex with whom? It's the same sex. How many partners were being changed? Were they using unprotected sex or whatever? And so

they started this survey, and they had it all planned, and then Margaret Thatcher took the money away. She said, oh, government, we're not paying for anything like that. So they found some more money from the Wellcome Trust, and it's been going ever since. Oh, I see. So that was essentially public health. It was a public health, and it's still public health. The people paying for it, the Wellcome Trust and it's been going ever since. Oh, I see. So that was essentially public health. It was a public health and it's still public health. The people paying for it, the Wellcome

Foundation and other departments are paying for it to find out about sexual behaviour because that allows you to know, to plan sexual health services. You were saying there about obviously a government where it seems to show a level of prudishness. Do you see in different countries a sense of greater

revelation? Because you would think that maybe in Britain there would be that kind of, I'll go up to the point where I take off my trilby, but the rest is very much between myself and my wife, whoever she may be. So do you see that there are some countries where you would think, yes, these are people very honest and that perhaps in other nations

you may be a little bit more dubious about statistics? Yeah, I mean, they've tried to do big national surveys particularly to do with with sexual dysfunction the pharmaceutical industry have tried to do this and found that actually getting in the far east and asian communities is actually very difficult to carry out interviews such as this in the states they they also had their big survey banned but now they ask these questions and it's the survey of families they call and so they've hidden all their sexual questions within this the survey of families, they call it. So they've hidden all their sexual questions within this national survey of family growth, and that's allowed them to carry this on under federal funding.

Paul, you're a man of the world. You've been both a mathematician, which I know on your website it says you don't like to talk about it, so we've talked about it on every time you've been on the show. It was a long time ago. You're also a comedian. In terms of what you've heard so far and in terms of your own experience, have you used mathematics more than humour

in the wooing process or the other way around? Well, I mean, there is the elephant in the room here, isn't there, that there may be... Obviously, there is mathematics behind meeting people, but if you mention to people that you are a mathematician, the chances of a sex session are incredibly low. So that's one of the reasons I don't talk about it. That was 20 years ago. 20 years ago, I did a maths degree, and I've been just living it down ever since. And that's why I went

into show business. I took extreme measures. So yes, in answer to your question, most of my sex sessions have been because I'm in show business. Or, I suppose, my own personality. There must be something genuine about me. So maybe people are attracted to that, but definitely not the mathematics. Are you disturbed that when you said, it's down to my own personality, you got a big laugh?

It's not fair, is it? No, well, I don't know what my personality is anymore. I've been in show business for so long. Hannah, in your book about the mathematics, love, you talk, for instance, about symmetry. And this is something that we've seen quite a lot in newspapers. Once every three months, they'll have a picture in a newspaper going, this is the face that everyone loves most. So can you run us through partly what that also means not

merely mathematically but I know that some of this is also kind of biologically revealing as well yeah of course so scientists have been trying for a very long time to really capture the essence of what it is that makes somebody beautiful and there are a few different things that sort of work and one of them is symmetry which is that people tend to prefer images of people with naturally symmetrical faces but the thing about beauty is that every time there's a rule there's sort of a counter rule if you like because while that works wonderfully for pictures of people's faces when it comes to moving images so videos or people in the flesh actually people tend to prefer asymmetry because it's seen as much more authentic.

in the flesh, actually people tend to prefer asymmetry because it's seen as much more authentic. Can I ask, what about if you have a symmetrical face but both sides are ugly? LAUGHTER I mean, it's less than ideal. It's less than ideal. We were talking earlier about the algorithm for approaching people at a party.

So that supposes that there's a... First of all, you can make a list in the model. We're talking about mathematical modelling here. So you can make a list of... I suppose there's a figure of merit you can attach to someone's attractiveness. And the suggestion is that people pair off according to the figure of merit you can attach to someone's attractiveness and the suggestion is that people pair off according to the figure of merit so I think you said something about the your chance of doing the

best that you possibly can yeah so so that is that borne out by the data that attractive people by some measure tend to pair off with attractive people unless attractive people tend to pair off with less attractive people well I mean it's slightly difficult to get really good data on that. One thing you can say, in terms of couples and how they match up with each other, because you have to have an objective third person measuring how attractive those couples were. But one thing that there is good data for

is how well or how popular attractive people are on things like online dating websites. So there's a wonderful study done by OKCupid, whose founders, incidentally, are mathematicians, so have deliberately built in bits to their website to allow them to experiment on their customers. Which takes... You're allowed to rate how attractive you think other people are

on a scale between one and five. And you can look at how that attractiveness, the average score of attractiveness, relates to how popular somebody is, so how many messages they receive each month. And you might expect that the more beautiful people find you, the more messages you'd get. That would kind of be the obvious thing.

But actually, it's not true at all. The thing that makes a difference is how much you divide opinion that's the thing that really counts so people who uh some people think that they're very attractive but other people think that they're very ugly indeed those are the ones who do very well so the people have something a bit strange or a bit quirky about them and and the suggestion or the reasoning behind this is that when you're looking at how attractive somebody is you're not just thinking um you're also thinking about your own chances of getting them so if you come across somebody who is generically attractive and kind of girl or guy next door you're imagining that they're going to be very popular online and that they'll get lots of messages and so yours will get lost in the thousands you think that there's a lot of competition for this person.

Whereas if you come across somebody, say, who's very beautiful but got lots of tattoos or lots of piercings or something a bit strange about them, which is your cup of tea, you imagine that there's a lot less competition. So it's a better bet for you to try and get involved. You talk a little bit in your book about... I don't know how much it's based on truth, but in the film Beautiful Mind, about the work of John Nash,

there is this certain element of kind of game theory that comes into when he goes out, you know, partying with... If John Nash ever partied, that doesn't seem like a thing, but with three other scientists. Can you give us a little bit about the game theory element? Yeah, there are a few different things. One thing that's quite important to remember when it comes to dating is that it's not like going to a supermarket and picking up the best looking

tomato, right? Somebody's got to like you as well. So there are a few different tricks and techniques that you can use to make yourself appear more attractive. You can use some tricks from mathematics and economics to manipulate how attractive people think you are to give you a better chance of sort of capturing the person that you like. So what things like, say, you've got more money than you have, pretending you're going to see them again next week, that sort of thing. I mean, lying would work, yes, for a short time. But one of them comes from something called discrete choice theory, which is about when you give somebody a number of options

and they have to make a decision. And in particular, something called the decoy effect. called discrete choice theory, which is about when you give somebody a number of options and they have to make a decision. And in particular, something called the decoy effect. Now, this is used quite a lot of the time in, for example, a very famous example of this would be in cinemas. So when you go to a cinema, there might be a small popcorn that's like, I don't know, £3.50 or something, and a large popcorn that's £6, which is ludicrously expensive. But on the list, they also then include a medium-sized popcorn, which will be like £5.50. Now, no sane person would go for the medium popcorn

when for 50p more you could have the large popcorn. So it's kind of an irrelevant alternative. But because it's on the list, it makes the large popcorn suddenly appear as though it's more attractive. Now, you can use this same effect, the decoy effect, to make yourself appear more beautiful. And this is an experiment that was done by an economist, Dan Urely, in the States. And he found two students of his that did a survey and that were considered equally attractive. So 50% of people said one was more attractive, 50% of people said the other.

And what he did is he wanted to create an irrelevant alternative, one that nobody would pick. So he took one of the men's faces and he used Photoshop to make them as ugly as possible. And then he sent out to 300 people these three faces, the two original faces and an uglified version of one of the boys. Now, nobody chose that uglified version. It was an irrelevant alternative. But its presence on the list served to make the original owner of the face appear more attractive. So suddenly, 75% of people thought that the person of the original face was more attractive. And if you switch over and uglify the other person on the list, the original face suddenly becomes more attractive. So the sort of result of this then is that

if you want to make yourself appear as attractive as possible when you pick a wingman or a wingwoman going to a party, you should pick someone who's as similar looking to you as possible, but just slightly less attractive. That's why I always go drinking with Brian. Yeah.

Little does my ugly friend Brian know. I mean, a relevant alternative alternative We did swap cardigans once, didn't we? Yeah, it changed everything It's really bizarre, put him in a cardigan and he looks as awful as me I would like to make it clear that when we first started working the show I had a full head of hair that was dark

and I've changed quite a lot and he hasn't changed at all suggesting he uses some kind of physics to steal my life force. David, one of the few chapters in your book we can talk about on Radio 4 is that you write about the probability of a couple... So we've met now, we've gone through, we've... We've gone through?

We've gone through? Deviously selected... We've gone through. We've gone through. Deviously selected a partner using Hannah's algorithms. So the probability of a couple having a boy or a girl, which is interesting. Yeah. And there are many nuances to the sex of your baby. Yeah, it's a subtle thing. Yeah.

For every 20 boys that are born, 21 girls are born. And this was recognised back in the 17th century. A guy called John Arbuthnot looked at data over 80 years and found that in every year... Sorry, more boys than girls being born. And he put it down to... His paper, he wrote on it. He said this was an argument for divine providence, the name of the paper, because he thought this must be the will of God

because of the increased mortality rate of boys, so that more boys would be killed in wars and things like that. In fact, he did the very first statistical significance test to show that the probability, if they were equal, 50-50, of getting 80 years like that, was one over two to the 80. So it's 21 boys to 20 girls. And the point is, though, that fluctuates. It actually, if you plot over the last 150 years,

you can see that there are peaks and troughs. And the peaks in the UK, for example, were in 1919 and 1944. So more boys are born at the end of wars. And this is reproducible. It's the same in the States, same in other countries. Why are more boys born at the end of wars? It's not a big peak, but there it is.

And people have argued about this for ages. And there's this lovely theory, which I like, which says that the probability of having a boy is related to what I shall term coital rate, is the formal term. In other words, how much sex is going on. And the argument for that is that during a month where the woman's got fertile in the peak in the middle, but if you're having lots of sex, sex every day, twice a day,

it's more likely that she'll conceive earlier in the month, and there's evidence that conceptions earlier in the month of a small tendency are more likely to be a boy. OK, so that's fine. So what happens at the end of wars? There's a lot of sex. You know, men coming home on leave, coming home being demobbed,

a lot of unprotected sex. The most babies that have ever been born in this country in number is in 1919. So there's the story. Frantic fornication breeds boys. The interesting flip side of that, of course, is to look at where the dip in the graph is. The dip in the graph was about 1900, when the fewest boys were born. And that had declined all the way through the Victorian period.

And historians have used this... It's said that, actually, what was happening in the Victorian period, why people weren't having so many children, was that they were not having sex. And the dip in the number of boys actually reinforces that idea that there was a lot of abstinence going on. Because Queen Victoria put them off.

Well, she relished sex. Yeah, she was all miserable, wasn't she? I suppose if you saw a picture of her over the bed or something, it would put you off, wouldn't it? So you're now saying that, depending on the attractiveness of the monarch, that will also happen. Who do you currently have? Which of the monarchs of England that will also happen. Who do you currently have?

Which of the monarchs of England, post-1066, do you currently have next to your bunk bed? Well, I always find a picture of Henry III is always what gets me going. Or Queen Anne. Lovely. Very elegant. You get lovely, elegant Queen Anne furniture, don't you?

Very elegant body as well. I think we're going to see quite a spike in the sales of images of Queen Anne amongst Radio 4 listeners. Can you describe Henry... If you could check that out statistically, please, David, in a month. Can you describe Henry III, out of interest? Well, Henry III. Very, very tall, but very gentle.

I don't actually know. Are there any pictures of Henry III? I don't know. You made up the Henry III. I believed you had a picture of Henry III above your bunk bed. Does anybody know what Henry III looked like? Of course I don't have a picture of him. A picture of Camilla.

It's not for sexual reasons. It's for loyalty to the crown. I was going to ask, Hannah, given what you've heard, how would your algorithm deal with someone like Paul? Well, he's definitely quirky.

There's that going for you. So, uh... It's an undone question. It was too much. Let's move on. So what, Hannah, thatannah that was apart from if we can just skip the bit about queen anne but we'll probably come back to it um where in terms of using mathematics to understand love to understand romance what are where's the shortfall what are the things that we can't analyze oh certainly there's a mysterious romantic something that that we can't analyse? Oh, certainly there's some mysterious, romantic something that you just can't capture with equations.

There's lots of studies that have been done to see whether you can write a checklist of things that you're looking for in a partner and whether that has any bearing on your long-term happiness. Once you actually get into a relationship, say things like, you know, what type of job they have, how much money they earn, what they look like, those kind of things. There's no evidence, really, no strong evidence whatsoever

that you can write down what you're looking for and it match up with what you actually want. But are people actually... I mean, that's the thing, isn't a lot of this, in the end, post-hoc rationalisation, where you think, oh, I was drawn to this person because of this, this and this, but in fact it was chemicals, which I know is not a popular kind of choice. One of those Valentine's cards, if you do just give a periodic table,

it's not there. But it's, you know, that... But there is... So a lot of the research and a lot of what we think we're drawn to is perhaps something of an illusion. Yeah, so there's the dating algorithms that you get on dating websites are trying to do exactly that, right? They're trying to take a list of things that you're looking for in a partner as well as your own characteristics and from the huge numbers of people available in the world to

try and filter them all out and deliver somebody to you who you're likely to have a meaningful relationship with very quickly. But the thing is that if they were really capable of doing that perfectly no one would ever go on any bad internet dates anymore right and i mean they certainly do and there's there's a couple of experiments that have tried to test out how effective these algorithms are there was one in particular where they tricked some people into thinking that they were more compatible than they actually were right they lied to them and said that they were a 30% match rather than a 90% match, and then tracked how that relationship evolved over time

compared to a couple or a number of couples who really were a 90% match. And there was a slight difference in how likely they were to exchange more than four messages, I think, the way they worked out was. But really, I mean, ultimately, the biggest thing about those kind of algorithms is that if you're told that you'll get on well with somebody,

you believe that you will. That's really the biggest thing. David, the use of statistics, as you said earlier, sometimes or often reveals counterintuitive results, as you said about the fact that the number of boys born after a war, shortly after a war, increases. And then there's interesting reasons for that you can explore. So I wondered what the most counterintuitive statistical result is that you've come across in this field.

Oh, I don't know about counterintuitive. The one that's constantly challenging is the conflict between a mathematical result and what you get when you ask people. And that's to do with asking people how many sexual partners they've had. Talking about opposite sex sexual partners. Now, in a closed group of the same number of men

and same number of women, the mean number of partners should be the same for men and for women. Not the median or the median, the mean number of partners should be the same for men and for women not the median or the mode the mean number because it's the both of those you know are related to the total number of partnerships you've got to when i do this in schools i talk about people shaking hands or whatever perhaps you should explain the terminology because we're an educational show as well education so what i mean is you know the the mean number if you number. If you've got a group of men and you add up, you say, how many sexual partners have you got?

And you add all those up and divide it by the number of men. That's the mean. And you do the same for a group of women, the same size. And as a closed group, those means must be the same mathematically. However, when you actually go out and ask people how many sexual partners they've had, men absolutely, essentially in every survey done, report having more sexual partners they've had men absolutely essentially every survey done report having more sexual partners than the women do in the usually it's double um but which is

mathematically impossible but in the last in the last survey for example among 16 to 44 year olds the average number for men was 12 and the women said they had an average of eight male sexual partners each which is just not feasible could not be the men having sex with each other oh no no no i know that's a good that's a good thing no this is opposite sex sexual partners so there's various explanations about this have they left out some women who have had huge numbers of sexual partners from the survey um people lying and the interesting one they've done, the experiment they've done on people to wire them up to fake lie detectors

and then ask them how many sexual partners they've had, and then the numbers start matching. The other thing is, are people just being a bit more, you know, are some people more careful about how they add up their numbers? Because when you actually look at the data, it's quite clear that after people have finished with one hand

or maybe two hands, it all gets a bit vague. They're going 15, 20, oh, 40, 100, 500 or whatever, except one man who said 47, which I think shows a remarkable grasp of history. So maybe... So there's a strong argument, maybe men and women, when they're looking back over what is possibly a rather ill-remembered history, actually make these guesstimates in rather different ways.

Could it be that if you've had a few wines, you could mistake the same woman for another woman? And put her again on the next tally? Quite possibly. And also, it's a suggestion that many women have had partnerships that they would just rather forget. But these are... As you say, these become interesting things to study

when you're talking about developing policy based on the data. So is it clear when... How do you design a questionnaire to try to get the most accurate data possible in these areas? People are struggling with this for years. The survey methods now are based on ensured confidentiality, face-to-face surveys, not online, but where the person being interviewed is answering them on a computer. The interviewer cannot see the responses.

The whole thing is locked down and shut off and made confidential. It's absolutely anonymised. Absolutely anonymised, yeah, clearly. It's a complete contrast to Kinsey back in the 1940s, very famous. He did 15,000 of his sex surveys. Back in those days, he made friends with them. He asked them in, gave them a cigarette, chatted

to them, you know, used vernacular terms, whereas now medical terms are used for parts of the body and actions. And he just chatted away and completed all these figures. He also used a policy of denial. So he would just ask people, when did you last have sex with someone other than your wife? Not have you, when did you last?

And they'd have to actively deny everything. So with the result that he got extraordinary stats that 50% of men had affairs, 37% of men had had homosexual experiences to orgasm, and 17% of farm boys had had sex with animals.

So this was in 19... I told you the archers was inaccurate. This was in 1948, so you can imagine the fuss it made at the time. What question, what loaded question did he ask to elicit that? When did you last have sex with an animal? What, and 17% of people folded and said last Wednesday?

Told them, told them when they last had had it. They just panicked. But could that be because... It was a long time ago! And also, why did he only ask the farm boys? I don't know, That was just 70%. In the rest of them, it was half. It was 8% in the general male population, 17% in farm boys.

It's a matter of access. Yeah, it's just less... When you're not a farm boy, it's less convenient, isn't it? You'll actually go to the farm, it's all the travel, both ends, you know, traffic, it's not worth it. But Kinsey's data are not considered hugely reliable. So what are the flawed areas of... In terms of, you know, Kinsey was used for quite a long time,

I think, as being a representative. What would you say are the statistics that are still brought up that come from Kinsey which actually now are flawed enough to dismiss them, or near dismiss them? Oh, I don't know, actually, because it depends. I think the same-sex behaviour ones, the 37% of men, because in a way it depends. He was using whole lives experience.

He was using sort of adolescent behaviours and things like that. And in the end, he came up with his figure. He's the one where the 1 in 10 figure came from, that was used so much in the gay rights debate in the 1970s. And that's been argued about ever since, this 1 in 10 figure. And, of course, now, with these decent surveys, you can get a very good handle on these. For example, sexual identity is now an official statistic

collected by the Office of National Statistics. Whether people say they're gay, lesbian or bisexual is part of official statistics. And that's a lot less. That's about 2.5% both for men and women would say they're gay, lesbian or bisexual. But if you ask about actual same-sex experience, what people have got up to,

then it's a hugely different answer. You know, now one in six younger women will say they've had a same-sex experience. One in 20, you know, has an actual same-sex partner. So the things have changed massively, but you have to be very careful what you define. Do you mean about self-identity or whether it's actual behaviour or whether it's attraction?

So three different things need to be very carefully deconstructed. So we're having two different discussions here. There's the measurement of behavior and then Hannah you're talking more about modeling of behavior so I suppose it comes it would come as a surprise to many people that we can even begin to consider modeling something as complex as human behavior in general particularly particularly sexual behaviour, using mathematics. So how well developed is that area of research? I mean, it's not... Well, you're right that it's a hugely complicated problem.

There are all sorts of things that come into it that you just can't possibly hope to capture with a mathematical model. But one of the things that I think is quite interesting is that through all of the noise, through all of the set of circumstances that led to you getting together with a particular partner, once you look at things from the level of a whole population, there are, as David's mentioning, these very distinctive patterns that come out, and patterns that have really elegant mathematical formulas. So certainly when it comes to looking at

sexual relations, the sort of results of the surveys that David is discussing, they have something called a power law relationship in terms of the number of sexual partners that people have. And this is something which hints at the way that the network of sexual contact is created. Essentially what it means is that you have something called a scale-free network, right? So you have these hubs, if you like, that sit at the centre of the sexual contact network, these people who've had a huge number of sexual partners compared to the majority of us who've only had, you know, a handful. And it's very similar to the way something like the Twitter network is structured.

So when you can draw those analogies between how our sexual structure or sexual connections are and other systems like Twitter then you sort of create these bridges between the universes that allow you to gain many more insights. So by power law relationship what do you mean for sexual partners let's say? Yeah so what it means is that the vast majority of people have roughly the same number of sexual partners, which is, you know, around the sort of 10 mark, I think, from those surveys between... Yeah, please, David. Well, yeah, I mean, the power law just means it's a massive long tail, a huge long tail, which it shows up in the data.

I mean, if you look at... If you ask this distribution I was talking about with an average of 12 or 8 or something for that, the mode, the peak of the distribution, is one. In the most common response, yeah, I've had one sexual partner. But then you've also got the people saying 1,000, 2,000, who are way off in the tail of the graph. Oh, I see, so that means...

And those really influence... They really influence the mean. In fact, the shape of the distribution might mean that the mean doesn't even exist, which it often doesn't in parallel distributions. But what this means is the really long tail on this distribution means that you have a non-zero chance of coming across somebody with any number of sexual partners, right? So it means that, you know, 50,000 sexual partners,

although there probably aren't very many of them in the world, if I've chosen a number that's too high, I don't know. 10,000. You'd have to be pretty old, wouldn't you? Mick Hucknall said he slept with 6,000 people't you? To be fair, Mick Hucknall said he said there were 6,000 people, and he's, well, Mick Hucknall. But I suppose this matters hugely.

If you're talking about health policy, for example, actually looking at the transmission of certain sexual diseases, sexually transmitted diseases, then where the mean is and where these tails are in the distribution matter a lot yeah because the spread of a disease is going to be influenced exactly by the network behavior that hannah's describing so it becomes absolutely crucial in in estimating you know what was a few years ago absolutely you know people were in a real panic about about the spread of hiv to the population and the old tombstone adverts and things like that.

So it was absolutely vital to understand what behaviour was going on and how that's changed over time. And now, you know, there's again a resurgence in sexually transmitted diseases. It's very important to understand, you know, how that's happening through what sort of network. How do you feel, Paul, about the listing to... As, you know as a former mathematician

who's left it a long way behind now for show business, the use of mathematics and statistics in understanding love? Because there are some people who I would imagine, they just don't like the idea. There are certain areas where the idea that science will get involved, it ruins the mystique,

as you were saying. How do you feel about that, Paul? Well, it's a numbers game, isn't it? I mean, the more people you ask, the more sex sessions you get. You just have to keep asking, and you keep getting more and more. As Plato said,

if you don't ask, you don't get. Well, exactly, that's how it works. I've met a... I don't know whether this really has anything to do with anything, really. But I met a gentleman in Denmark, and I said... He wasn't a particularly attractive gentleman, but he said that he'd been in many, many threesomes with various women.

He liked being in threesomes. He didn't like just one, he preferred a threesome. And then I said, How did you get into all these threesomes? And he just said, I asked. And nearly always I said yes. It's a tip.

So that's what I think. You've just got to keep asking. That was nothing to do with the question I asked. I wasn't glad you went down there. I can't really remember what the word was. No, it was a long time ago. It doesn't matter. It's like Gardner's Question Time. Full of useful little tips for everyday life.

But what for you, Paul, has been the most, of what you've heard this evening, is there anything that you would consider kind of revelatory from either David or Hannah about this understanding or these statistics? Well, yes, there is something I find revelatory. Maybe I shouldn't say it, but I can't believe in this day and age that the most common thing is people say

they've only had one sexual partner. And I can't believe that the mean is only ten, because people are all... Certainly young people, is that to do... Is that older people skewing the statistics? The point is that you get these changing behaviours, not just a function of age, it's a function of when you're born as well,

because you get these cohorts moving through and they take their behaviour with them. So various practices, which I won't go into in detail, now are becoming more common in older people because they've grown up with them. If you see what I mean. That was the enigma code of answers. It takes a while to crack that one, David.

Is it whips? No, no. No, I don't think we should start guessing. I don't think that's... So... This is very much reaching its natural conclusion of the show isn't it the um is there a way you think people could be happier using mathematics using statistics in terms of is there a practical way that those the people listening to this program

who are thinking maybe now maybe now i've, you know, the armoury of statistics and mathematics. Tonight is the night to go to that bar. Do you think that... Is that really true or is it just going to be something that's good for a book and a TED talk? This is aside from their natural love of the mathematics, right? Yeah. I think what's quite nice about using maths in this way, you have to sort of take it with a pinch of salt. I'm not suggesting that everybody should live their life by these rules. But what comes out time and time again,

when you look at things from this perspective, are quite positive messages, right? It's like, be proactive, go for the person that you're after, play up to whatever it is that makes you different, don't allow yourself to be exploited. I think all of those messages are generally quite positive, and things that you can forget when you're sort of stuck in the drudgery of being single for a long time. So perhaps not in the individual case, but I think the overall message is quite positive.

And I think looking at stats is quite positive because if you think, you know, everybody else has a massive amount of sex and all these partners and all this activity and all this time, when you actually see yourself where you are on the distribution of things, they make me think, oh, well, you know, I'm fairly normal. You know, the median number of times a couple now,

a young couple has sex over four weeks is three. It's gone down over the last 20 years. As I said, the most common number of partners people have had is one, although there's a very long tail beyond that. Many young people have sexual problems, as many as older people as well, who report having sexual problems. So, you know, it's not that every... You can actually... Many young people have sexual problems, as many as older people as well, who report having sexual problems. So, you know, it's not that every...

And that is reassuring to find out that actually, you know, maybe you are just sort of normal. I had a thing a bit like that, because... LAUGHTER This actually happened to me. About four years ago, I was sitting on a plane in Australia and talking to another comedian, and he'd revealed that the

night before he'd been on stage and sort of flirted with someone in the audience and then ended up having a sex session. So I thought, oh, you know, wish my life could be like that, you know. So then I just went on a massive sort of sexual rampage for about three

years. And then I saw him last year, also in Australia, and I said, oh, you know, I've got the life that you have now. I have all these sex sessions. And he said, no, no, that was just a one-off thing that happened three years ago. And he'd been really quite lonely for a long time. That is an actual thing that actually happened.

But it was based on a misunderstanding, because I thought, oh, that's what it's like in show business. You can have sex with loads of people. So you're one of those high-density nodes. I'm definitely a high-density node. An outlier. You're a long way down the tail, as we've discovered. So I was going to finally ask you,

but I was going to go, have you got any advice for anyone? But I think you've just given it. And David, for Radio 4, just that bit of going, young people have as many sexual problems as old people. It's made a lot of Radio 4 listeners go, good. So I think that's been an upbeat point as well. We asked the audience a question as well, and that question was, how would you use science to attract a partner?

And the answers are, I would eat as many deep-fried Mars bars as possible, eventually creating strong enough microgravity around me to attract a suitable mate. That's just a very lonely man. Which I also like, he's done Mars TM as well. That's the kind of accuracy we enjoy amongst our audience. This one seems like a direct

observation of Paul, actually. He says, act chaotically to become a strange attractor. Act like a quark and use my charm as a niche joke. I'd invent something to make me sound like Brian Blessed and look like Brian Cox.

Oh, God. His voice came out. Isn't it wonderful? I want to go to Mars! You couldn't do it. Your shows just, well, one, they could be over a lot quicker. Brian

doesn't do long, lingering gazes. Let's just get up the volcano and eat it! I would invent something to make my husband... Does that sound like Brian Blessed and look like Brian Cox? Another one there. Why are there two of those? They may well be married.

This is... I really have travelled back in time for the future to become Brian Cox. Hello, David. Show them your Brian Cox. Anyway, so the... LAUGHTER What have you done to us?

APPLAUSE Much of this will make the podcast version. Much less will make the 4.30 Radio 4 version. So, thank you to David Spiegelhalter, to Dr Hannah Fry and Paul Foote. Now, I hope you found this show practical and useful. And if you attempt to put into practice anything you've heard today and it leads to love, please do not write in and give us the details.

It's disgusting. Anyway, so, warning, by the way, warning. Prolonged use of mathematics to find love may lead to headaches, drowsiness and astonishing revelations about the shape and structure of the universe. Please consult a doctor if your view of the universe becomes unstable. On the other hand, ignoring all reason and deploying a purely illogical approach to your everyday life... Give John a chance to throw it away. In the infinite monkey cave.

I know she's not true. In the infinite monkey cave. That's just how it is. In the infinite monkey cave In the infinite monkey cave And then we've got the podcast, podcast, podcast, podcast version of the podcast, podcast, podcast. Brian doesn't even know that you have actually now listened to the whole of the show. And this is all he's been doing for the last 47 minutes. And it's not going to end for a while either.

It's a nested infinity of podcasts. You could probably sum it up. And this is all he's been doing for the last 47 minutes. And it's not going to end for a while either. This is my life. It's ooey, gooey, and just five bucks with a small coffee all day long. Taxes extra at participating Wendy's until May 5th. Terms and conditions apply. In our new podcast, Nature Answers, rural stories from a changing planet, we are traveling with you to Uganda and Ghana to meet the people on the front lines of climate change. We will share stories of how they are thriving using lessons learned from nature. And good news,

it is working. Learn more by listening to Nature Answers wherever you get your podcasts. you