#838 - Dr Paul Eastwick - What Do People Really Want In A Partner?
01:09:25 Share

What's happening, people? Welcome back to the show. My guest today is Dr. Paul Eastwick. He's a psychologist, professor, and a researcher. What do people actually want in a partner compared to what they say they want? Paul is the lead author on one of the largest studies of its kind, which was just released, breaking down exactly this question. Expect to learn the number one trait people actually look for in a partner, how well people know what they want

what ideal partner preference matching is, the biases that affect mate evaluation, the sex differences in stated versus revealed preferences, whether big data could improve dating app matching, and much more. A lot of uncomfortable insights today as our

very publicly acceptable proclamations about what we want, supposedly, and a partner are ripped away from our eyes and the harsh reality comes in to smash us in the face. Really interesting, fascinating stuff. The study is massive and very impressive. And Paul's great. So I really hope that you enjoy this one.

I've worn Whoop for over four years now, since way before they were a partner on the show, and it is the only wearable I've ever stuck with because it's the best. It's so innocuous, you don't remember that you've got it on, and yet it tracks absolutely everything 24-7 via your wrist. It tracks your heart rate, your sleep, your recovery, all of your workouts, your resting heart rate, heart rate variability, how much you're even breathing throughout the night,

puts all of this into an app and spits out very simple, easy to understand and fantastically usable data. It's phenomenal. I'm a massive, massive fan of Whoop and that is why it's the only wearable that I've ever stuck with. And best of all, you can join for free, pay nothing for the brand new Whoop 4.0 strap, plus you get your first month for free and there's a 30-day money-back guarantee. So you can buy it for free, try it for free, and if you don't like it, after 29 days, they'll give you your money back.

Right now, you can get all of that by heading to join.whoop.com slash modern wisdom. That's join.whoop.com slash modern wisdom. This episode is brought to you by 8sleep. I have been using my 8sleep mattress for years and I literally cannot imagine life without it. Having an actively cooled and heated mattress is the best.

the game changer. Now they've launched their newest generation pod, the Pod 4 Ultra. The Pod 4 Ultra can cool down each side of the bed up to 20 degrees below room temperature, keeping you and your partner cool. And for those of you who snore heavily, it can detect your snoring and automatically lift your head a few degrees to improve your airflow and stop it. This is your 3000 stuff. It cools, it heats, it elevates, and it is clinically proven to give you up to one hour more of quality sleep

This episode is brought to you by Element Sports.

Stop having coffee first thing in the morning. Your adenosine system that caffeine acts on isn't even active for the first 90 minutes of the day, but your adrenal system is and salt acts on your adrenal system. Element contains a science-backed electrolyte ratio of sodium, potassium and magnesium with no junk, no sugar, no coloring, artificial ingredients, gluten fillers or

any other BS. It plays a critical role in reducing muscle cramps and fatigue whilst optimizing brain health, regulating appetite, and curbing cravings. It's how I've started my morning every single day for over three years now, and I absolutely love it. That orange flavor in a cold glass of water first thing in the morning is fantastic. It is the best way to start the day.

Also, they have a no BS, no questions, ask refund policy. So if you do not like it for any reason, they will give you your money back and you don't even need to return the box. That's how confident they are that you love it. Head to drinklmnt.com slash modern wisdom to get a free sample pack of all eight flavors with your first box. That's drinklmnt.com slash modern wisdom. But now, ladies and gentlemen, please welcome Dr. Paul Eastwick.

Your new study is one of the most interesting things that I think I've seen this year. Oh, good. Oh, good. It also has maybe the highest number of authors that I've ever seen on a single paper. It's a big team. Yes. So how well would you say people actually know what they want in a romantic partner? Um, it's a big team.

It depends. So what people are very good at recognizing is that some attributes are very, very desirable.

right? So there's good agreement that traits like attractiveness and intelligent and considerate and honest, that these are desirable things that we want romantic partners to have. And there's also a lot of agreement that we don't really want somebody who's disorganized and careless, and we don't really want somebody who's anxious and easily upset. So there's a lot of agreement and accurate self-knowledge

That some attributes are more desirable than others. And you can ask lots of interesting questions about why don't we want to be with anxious partners, right? Why don't we want to be with partners who are kind of a mess, right? And why do we want to be with partners who are attractive and intelligent?

The trick, though, is when we expect people to have insight about what it is that they uniquely like, what do they like that makes them different from other people?

And that's the insight challenge where we find sometimes people do okay and sometimes not so much. Why is that an interesting insight? What is it that you like that other people don't like? Why is that an interesting question? Well, I'll tell you, the reason I got interested, the way I came to that particular question was because of

The work on gender differences. So what do men and women want in a partner? Right. And so this is research going back and it goes back 80 years at this point. Right. I mean, middle part of the 20th century when we started asking people is actually the sociologists at first were really interested in.

what attributes do men and women say they like and do we find these gender differences and you certainly do for attributes like attractiveness for attributes like earning potential right men will consistently say they like attracting this more than women women will say they like earning potential more than men um so we were originally interested in

whether we saw that those gender differences also played out when we looked at how those attributes predicted all sorts of downstream consequences, because that is an individual difference of sorts, right?

How do you mean? Well, gender, right, is what we're doing is we're describing how some people are different from other people, right? In some ways, it's like one of the easiest ones to latch onto in the mating domain, right? But it does function like other individual differences in that if men say something as a group that this appeals to them more than this other group, women,

It requires some amount of individual predictive power, right? That the groups have to be telling us something different that's going to then play out when we see what it is that they actually find appealing. So it was really the gender differences that got me interested in this accurate self-knowledge question in the first place.

Right. So we have two things that are going on here. One is what do people say that they look for? Yeah. And the other is what do people actually want? So we have stated and revealed preferences. Can you explain how you looked to sort of pull these two apart? This is the internet, especially online sort of

mating discourse. It's the favorite. Don't trust what people say, watch what they do. You know, it's the death of every evolutionary psychology survey or self-report that's ever been done because it says, well, no, no, no, that's what people want to say. So talk to me about how you tease these two things apart and avoided sort of too much confabulation between the two. You bet. And this is a key distinction. And they're

Yeah, these terms get bandied around, but I'll tell you how I use the terms. And I think this is a very helpful way to think about it. So when we're talking about attributes, um,

A stated preference for an attribute is usually done very simply. Here are a bunch of attributes. Here are some rating scales. Tell us how much you would like these attributes in an ideal partner. You can be more specific. You can say ideal short-term partner. You can say tonight. Sometimes we would do studies where we'd say, you know, oh, when you go speed dating, how much are you going to care about these attributes? But all of those fit under the stated preference rubric. It's, you know, I see this trait and how do I feel about it? Okay.

A revealed preference, it's not, sometimes people confuse it with like, oh, but what do you actually choose? That's not actually exactly what it is. A revealed preference is about what does the attribute predict for you? Okay. So if you meet an array of people who vary in that attribute,

Does that attribute help distinguish the people you liked from the people you didn't? Okay. So if I send speed dating is a very helpful way to think about this, even though you don't actually need speed dating to get revealed preferences, but it's, it's helpful. I find because you can imagine meeting a set of 10 or 20 people and some are very attractive and some are a little attractive and some are not attractive at all. And again,

The extent to which you have a revealed preference for attractiveness is the extent to which attractiveness is a driver of the liking that you experience for these people. Now, liking could also be like a self-report scale. Liking could be a choice you make. Liking could be like, who do you take on dates, right? Liking can be measured in a million different ways, but it's some sort of association, some sort of predictive relationship between the attribute and the

some sort of evaluative experience that you have for a set of potential partners.

What did you do to work out the, I understand how you can do stated preferences. You just give people a report. You say, here's a list of 35 traits, rank order them from one to 35 in terms of which one you think is most important. How do you discover 10,000 people's revealed preferences? Right. So there are a couple ways of doing it. And we actually did it in two different ways in this paper.

The main way that we did it is we just actually look at the revealed preference in the sample. So in the whole sample of 10,000 people – and I should probably explain briefly what we did in the study. We have a survey. It's about 10,000 people from 43 different countries.

And some of them are in established relationships. Some of them are single. And what they're doing is they're going to be reporting on somebody that they're like kind of interested in dating. Right. But but they're not dating currently.

And they're completing a bunch of traits about these people, right? They're rating them on 35 different attributes. We also have their ideals for those 35 attributes. So we know how much they say their stated preferences for those attributes. And we've got a dependent measure to have, you know, sort of how positively do you feel about this person, right? It's sort of sort of standard liking desire kinds of measures.

So what we can do is in the whole sample, we can just look at, okay, when people felt that these partners were, let's say, a good lover. I'll use that example first.

How positively did they feel about this person? And it turned out that a good lover was the single strongest predictor of that dependent variable, right? So a good lover is a very strong predictor of feeling positively about a romantic partner, right? That relationship was the strongest. So we would say in this sample, there is a very strong revealed preference

for a good lover, even though people ranked it something like 12th in terms of their stated preferences. It was their number one revealed preference. So that's, I think, a helpful way of thinking about this distinction.

What is ideal partner preference matching? I know that this is kind of a body of work. Just I feel like that's important to get it out there. Yeah, right. So the one of the questions we can ask in this literature, I mean, looking at the stated reveal, it's all very interesting. But sometimes we want to know, OK, if you are a person who says you like attributes X, Y, Z.

Do you like partners more when they match rather than mismatch X, Y, Z? Okay. So if you say you're somebody who really wants somebody who's attractive and intelligent and funny, you really don't care about attributes like, are they religious and successful? You don't care so much. Okay. Does the extent to which the partner matches those ideals, those stated preferences that you have, does that predict how positively you feel about this person?

And figuring out the right way to do that matching has been...

Very complicated. And it's actually taking the whole field like a decade to sort out how to do it. Because it turns out there's all sorts of complex stats that go into creating that matching process. Anybody who's ever played around with things like profile correlations or different scores is going to be familiar with the many complexities that come into play when you try to do this.

But what we tried to do in this study was basically take all of the different approaches that are out there and say, hey, we're going to do them all. And we're going to show you how well these different matching approaches work to try to get at the question of whether people are happier, more desirous of partners who match rather than mismatch their stated preferences. And what did you find?

Well, what we found is that if you pull out that unique component, okay? And again, this was, we really didn't know what we were going to find going in, what that unique component was going to show. Because I'd seen studies where this component was basically zero.

Um, and, uh, and, and other studies showed things that were a little bit larger than that. So what we found was that when you pull out that unique component across all 35 traits, we found real effects. They're not huge. It's like two, three, 4% of the variance tops, but it's there. So a matching effect.

that is truly about how much you desire these attributes. We've taken all of what we call normative matching out of it. We just look at that individual differences component. So if you meet somebody who uniquely matches what you say you're looking for, you'll experience more desire for that person. It's not a huge effect, but it is there. We are able to detect it. Does that say that we have some degree of insight into what it is that we want in a partner? Yeah.

Yes, with the major, major caveat of when we're thinking about collapsing across 35 attributes. So across that whole set, we get a little something.

The challenge is, what if you care about one attribute? What if you have a hypothesis about attractiveness? Or what if you have a hypothesis about intelligence or warmth? What happens then? And that's kind of where things start to fall apart. Why?

Well, those effects, because there are other ways of looking at matching effects if you care about single attributes and isolation. And this is a basic statistical interaction prediction. So again, for listeners who are familiar with these kinds of stats, what you're talking about is if you're the kind of person who says you really care about attractiveness, okay? You rate that highly in an ideal partner, that's your stated preference. We want to see...

Whether it matters that for, again, how much you're desiring this person that you think that person is attractive. So you should experience a stronger association between how attractive you feel someone is and how much you desire them. If you say you care about attractiveness. Okay. So that's the way we test that prediction. These are statistical interactions and you do it for all the 35 traits.

And generally speaking, those are tiny. Because the sample is so big, we detect many of them are significant samples.

Although many of them are not, and actually sort of shockingly, you know, like attractiveness isn't. But some are, and some are big enough to be notable. Actually, the one that really stood out was religious. So I would say tentatively, like if you're the kind of person who says you want a religious partner, there may really be a matching component to that. But for most of those attributes in isolation,

Those effects are very, very small. I mean, you need like thousands and thousands of people to be able to see them. Okay. Well, I mean, the least interesting part of that study is people were able to slightly predict the things that they want. The most interesting part is what is it that people stated that didn't come out as the revealed? So can you take us through where people really missed the mark and sort of what you think is going on with the motivation for those?

Yeah. So there are some fun discrepancies there that surprised us a little bit. We didn't, you know, I don't think we'd ever run a study where we had 35 different trays where we felt comfortable doing this kind of ranking approach where we take all

all the revealed preferences, kind of stack them up. And we take all of the stated preferences and stack those up and we sort of see where the matches and the mismatches are. And so I mentioned one of the mismatches earlier, which is that people really seem to have a strong revealed preference for a good lover, even though that kind of ends up being about 12th in their stated preferences overall.

There are a few other interesting ones. People don't say they care that much about an attribute like smells good. That kind of fits somewhere in the middle of the range. They don't spontaneously think that sounds particularly appealing, but it actually ends up being the fourth most

biggest thing overall uh in terms of uh in terms of uh what they feel uh their revealed preferences for it and then of course if some are if people are underestimating on some they're going to be overestimating on others and they overestimate it on some things like uh oh uh like consider it a little bit uh other attributes you know that are um

A little bit, you know, more on the on the warmer side. People, you know, say that those attributes are very, very important. But then when you look at the revealed preferences, it seems like a patient patient. Yeah. Patient number 10. And it comes in at number 18. Yeah. Interesting.

um what else have we got here the the one the emotionally stable i think was another one that like yeah people say they really want somebody who's emotionally stable but i mean it's it's positive but it's it's not as as strong as some others so so maybe you know people in the end are okay with partners who you know maybe have a what's this one going from 19 to 6 here sexy yeah sexy from 19 to 6 okay so this is interesting so actually with

a lot of the attractiveness related traits, people were underestimating those on average, right? Sexy was one, I think nice body was one. And so to bring it back to the gender differences, we can look separately for men and women at what their stated preferences show and what their revealed preferences show. And on the revealed preferences side,

what you see is that men and women are really getting these attributes about the same. Okay. Attractiveness, sexiness, nice body. Those have the same revealed preferences for men and women. And that, you know, does that shock people in today's day and age? It doesn't shock me because this is what we've been seeing for like 15 years. When we look at speed dating or we look at

ongoing relationships, revealed preferences for things like attractiveness really don't show gender differences. The stated preferences, of course, do. And what we can show with this design is that what's happening here is that both men and women are underestimating

how much they like attractiveness, but women are like really underestimating. So, so the underestimation effect is bigger for women than it is for men. Right. So, so that's how they both end up at the, just to dig into that. Um, men prioritized in revealed preferences. Yep.

And in stated preferences? Yeah. Attractiveness more? Yes. The men and the women prioritize, in terms of revealed preferences, attractiveness equally. Right. But on their stated side, the men prioritize it more. But the men are still underestimating. Right? So the men are still too low relative to other traits in terms of how they rate the stated preferences compared to where the revealed preferences come in. The women are like way too low.

What else was interesting when it comes to the sex differences between men and women? So we can also look at some earning relevant traits. We've got things like ambitious and we've got things like a good job and we've got, you know, attributes in that space. Financially secure is another one.

And so those as stated preferences generally rank quite a bit lower. But you see the gender differences that men that women say they care about these attributes more than men do. Once again, on the revealed preferences side, it's the same for men and women. And again, we've been seeing this for for 15 plus years now.

But again, what's interesting here is now we can figure out, well, who's getting it wrong? And this is a case where actually both men and women are getting it wrong, just in opposite directions. So women are, in their stated preferences, are overestimating a little bit how much they like those attributes. And men are underestimating a little bit how much they like those attributes. They aren't dramatic errors.

But they're big enough and in opposite directions that it explains why you see a modest gender differences, gender difference in what men and women say they want on those attributes and how they end up at the same place in terms of their revealed preferences. Why do you think it would be the case that women would overestimate how important a good job is in a partner? I would have guessed that.

The opposite. I would have guessed that it's a stereotype about women. They would have compensated for it. They don't want to seem like a gold digging hoe. So they're going to counter signal against that. And then they're going to be smacked in the face by the reality of being a safety and resource seeking person. But it doesn't seem to be the case.

Yeah, you know, I actually think there's some evidence that the gender difference and I think the overall preference for those kinds of attributes that they have gone down a

a little bit over time. I could be wrong about that, but I think that makes sense when you think about it in terms of, as women have entered the workforce in greater numbers and entered more high prestige jobs, you're likely to find that at least in terms of stated preferences, they're not thinking in the same way that they would have back in the 50s. Like, oh, I need a

partner who's going to really be making the money because I'm not going to be earning as well. So I do think the stated preferences have come down in some sense, but I also think you're right that some of what is affecting people's stated preferences is

are stereotypes. And I mean that in the most neutral way possible. Just when you ask people to describe the attributes that another group has, right, they can do that. And I think some of what men and women do when you ask them to describe their ideal partner is they start picking out attributes that members of the other gender have. And that's the funny thing is that men and women know that men tend to

earn more than women, and attractiveness works the other way. So men and women know that women, on average, tend to be more attractive than men are. So there are some of these, I think, stereotypes, again, in the most neutral way of using that term, kind of infect the way people provide those stated preferences in the first place. And that could be a source of some of these discrepancies between what people say they want and what they actually want.

Yeah, I'm sort of trying to work out what's a black pill and what's a white pill from the study. I think it's really interesting, right? Because this is, I'm going to guess, the largest study ever done on stated versus revealed preferences for men and women. It's pretty close, yeah. Okay, one of at least the most recent. It's hot off the press. Yeah.

Just for the people that are only listening, well, I've had the table up on screen for the people that are watching. A good lover coming in at number one, at number two, loyal. So, you know, we have this big discrepancy in a good lover. We have this big discrepancy in smells good and sexy. So it seems like people wildly underestimate the importance of some more kind of slightly shallow physical characteristics that are to do with intimacy, at least like intimacy.

Physical intimacy. But then when you look at the rest of the top 10, you've got loyal coming in at number two. You've got honest, understanding, considerate, supportive, you know, a lot of much sort of softer people.

traits, sympathetic and warm. You know, these are so we have sort of two things going on at once, like the red pill and the blue pill are both kind of right at the same time here. You know, you've got this vicious purple pill. Yeah, it is. It is a purple pill.

But it's this real blend. It's a real blend of the two. There is a sort of a focus on both the immediate attractive sexiness and the softness too. I think that's right. And again, what we're seeing here is that, you know, these are people's

own judgments, right? So I'm making judgments about this person I know and whether I find them attractive and whether I think they're a good lover, right? So, so we're, we're really getting at an individual psychology about the kinds of traits that really co-vary very strongly with good feelings about that partner, desiring that partner and wanting to be with that partner. Um,

There are other lessons. And again, I'm now drifting away from this study a little bit with respect to, okay, but if I'm trying to attract somebody in the first place, I think that's where in many cases people can find the red pill pretty frustrating. Cause if you feel like, well, I'm not attractive or I don't have these traits that make me seem like I'd be a good lover. I'm at this major disadvantage. And again,

My usual response to that is like, look, in a setting where people are meeting for the first time, that is true. People reach consensus about attributes like attractiveness or being sexy, being confident. These matter in a first impression context, but.

Where the blue pill comes in a little bit is that it says, but as people get to know each other, if you're in a context that is going to create some repeated interaction, consensus on those kinds of attributes tends to go down. So over time, we start to disagree more and more about who's attractive and who's not.

So for somebody who's not conventionally attractive, for somebody who doesn't go into the party and, you know, attract all the attention, what it means is that there are other avenues for you, but they have to be avenues where you actually get to know people over time because some people's opinions will diverge. What do you make of conventional and uncreative being ranked at number 35? What do you think that says about people?

Yeah, that's a great point. People certainly like to feel some excitement, right? They want to feel like their partner is going to inspire them. And this is very true in today's day and age. It has become more true over time.

that we want our partners to push us. We want our partners to help us like become closer to ideal, our ideal self means like a core idea in, in close relationships research is that we take on attributes of our partners and we want to experience new things with our partners. So

who is not terribly creative, that doesn't sound like a recipe for a sort of a fulfilling, you know, growth-infused relationship. Absolutely.

I wonder because it's sort of trustworthy is one of the things that's in there. And you think what trustworthy is kind of like, it's tangential to that. It's this sort of reliable, predictable kind of way. But we, it's, you know, this is exactly where the devil's in the details of we want predictable, but not too predictive. I don't want them to be reliable, but I don't want them to be, you know, to be able to know what they're going to say every single time I have a boring conversation with them. So it's this, it's this sort of delicate balance between the two. Yeah, right. Right. And we like people who are

who we find exciting, right? Who are adventurous, who are going to do new things with us. But we also don't want, you know, close relationships are by their nature, they're dangerous, risky things. And they're dangerous and risky things because you put yourself in a position to be taken advantage of, right? I mean, that's kind of the whole point. Right.

is that you're establishing intimacy with somebody. You are opening yourself up. There are tremendous benefits, both psychological and tangible, that come with that. But...

It is a risk of exploitation. There is a risk of being rejected, of being taken advantage of. And so this is why I think things like loyal often end up, you know, pretty high on the list because people want those reassurances from somebody, but they want the excitement to just excitement that isn't going to like, you know, mean that that ultimately you're going to reject me.

Yeah, not excitement. I find you in bed with a different person. That's right. That's a little too exciting. I totally didn't think about the fact that people take on the traits of their partner. The fact that I didn't think about it probably means that other people also didn't think about it. Sure, sure. But you are going to absorb some of the traits of your partner. So if they are anxious, it's not just anybody that's ever been in the car, any...

that's ever been in the car with a very nervous co-passenger girlfriend. Oh, no. And they panic about traffic. And you're like, I'm a cool driver. You're making me panic. Like, stop. I'm just trying to do my thing here. And you kind of absorb, even if it's just for a brief moment. But that's one of the reasons. Like, if it's somebody who has an anxious disposition, if it's somebody who...

The most literal way, if it's someone who smells bad, not only do you have to smell them, but you probably smell bad too. Yeah, right, right. You take it on yourself. And this is, you know, it's part of why this work is called the sort of self-expansion and inclusion of the other in the self. But one of the reasons that researchers think that initiating relationships can be so exciting is

is because of this self-expansion experience that you get when you're spending time with somebody new that you're really excited about. Because ultimately, you're introducing each other to new things, new activities, new interests,

new ways of seeing the world. And it's all very invigorating and very exciting and part of the reason that you don't need to sleep as much and you don't need to eat as much, right? It's all very self-expanding in that way. How much do you think the ranking that people have given when it comes to their revealed preferences

of particular traits is simply choosing to not have the absence of that trait. So the selection of good lover being the selection of not a bad lover, the selection of smells good being the selection of doesn't smell bad. What do you think about that? That's a great question. You know, we have occasionally looked at

for those kinds of effects, but it's very hard to tease apart with these kinds of data, right? But because what you might be seeing is that if I think somebody is like a one or a two or three, I am penalizing them exceptionally harshly. And I'm not really differentiating between the five, sixes and sevens. Okay. Um,

When we're doing studies like this, where people are reporting on it's either somebody you're in a relationship with or somebody you're kind of into all of this, you know, these differences that we see here are really happening in that upper end. Right. Because this is about differentiating somebody who I think is a seven on attractiveness from somebody who I think is a six.

But in context where you're meeting strangers and it's maybe a broader set of possible people that you'd be meeting, I do think it's plausible that people are sort of establishing some minimums and working up from there. But it sounds like...

very clear and easy to picture how that would work. The stats end up being absurdly complicated. Yeah, I imagine they are. The stats was actually one of only two subjects that I got a C in at GCSE level. So I was early onset bad at statistics. Yeah, it's tough. Thinking about the good lover one, which is, I think, not only the starkest difference between stated and revealed, but also then comes and ranks at number one.

What do you make of that? Do you just, is it a, um, both as the gap and then as the fact that it is the revealed? That's a good question. I think that when people, when you imagine somebody who's a good lover in the abstract, that what people are probably thinking is, um,

that's not going to impact my day-to-day life. What are we talking about? We're talking about 30 minutes every evening or 30 minutes every week if you've been in a relationship for a while. But then in reality, somebody who's a good lover, that idea of

incorporates a lot of other good things that relationships have, right? So it also indicates things like they're sensitive, they're giving, they're caring, right? But they also know what they're doing and they care about whether my sexual experiences are good, right?

So I think it's one of these attributes that when it's disembodied and abstract and disconnected from a particular person, it sounds just kind of okay. But when we think about our partner that way, it sort of brings in all of these other components.

that really make a sexual or romantic relationship what it is. I imagine that's a really good point. I imagine that smells good, has a good body, is sexy. What is it? It's somebody that has a good body thing, reliable, disciplined, has agency over their life, they're motivated, they can overcome discomfort. It's like this whole list of things that are upstream from...

has a good body yeah and i guess the same thing goes for smells good it means that they've probably got good personal hygiene and they take care of themselves and maybe they're considerate of others and so on and so forth you know another this is just occurring to me now and this would be an interesting thing to test is that when you ask people how much do you want these traits in an ideal partner they are mostly thinking about um

you know, who I want with me, you know, by my side doing the hard tasks of life. But when they're thinking about a real person that they are in or want to be in a relationship with that, these sexier, more romantic, um,

These components come to the fore more. It's a bit more visceral. Yeah, right. Because I'm thinking about this person and all the visceral things that they inspire for good or for ill. And so I think that's that's quite that's another possible explanation for why we see some of those discrepancies.

Were the stated or revealed preferences for short term or long term relationships or both? These were all we usually describe it as an ideal romantic partner, which at least to me connotes something long term. We didn't have we didn't ask about short term ideals in this study, really just, you know, because of time.

Got you. Did you work out overall whether men or women were more accurate? That's a good question. Actually, I don't think we did. But that's something we could certainly do. That seems like simple stats. Yeah, right. That shouldn't take too long. Right. That shouldn't be too bad. My sense is that they're probably similar, but I don't...

But I didn't look at whether there's sort of overall mismatches there. I mean, certainly we saw with attractiveness, women had it off more than men, and for earning potential, they were kind of similarly off. I wonder, did you consider or would you consider doing a follow-up study to do the reversal? How important do you think that the opposite sex thinks that trait is in you? Yeah, that's an interesting thing. And, you know, I mean, there's some studies that have done that not at this kind of scale.

But people's ideas about what the other gender likes, I think the evidence suggests it tracks pretty well what people say they like, right? But we've never tried to look at, well, what would the discrepancies be there, right? Between my ideas about what you like and your revealed preferences for, you know, somebody of my gender. Because that's what's important. And this is, I think, probably the most interesting element. Basically,

Is it even worth talking about mate preferences if they don't actually reflect our revealed mate preferences? Like what we're both doing is we're both playing this game. I can predict what you say you're going to want, but that's pointless. I'm not here to just play some lexical game. The reason that we try and predict what other people want is to predict what they want, not to predict what they say that they want. Right, right. I do agree that this is part of the challenge here. And again,

It is true when we talk about that whole suite of 35 traits that we can get some action there. They're not miles away. Right, right, exactly. That people do seem to have some self-insight. But on the specific traits, usually my...

bias as a researcher is that I'm usually much more interested in those associations, what we call reveal preferences here, that I want to know how strongly does this attribute predict some DV that I care about?

I think people's ideas about the attributes they like are also interesting, but not because I'd want to use it strategically to try to match it exactly. I'd want to know what their revealed preferences are and try to match that. Are you familiar with the Keynesian Beauty Contest? Do you know what that is? No.

No, tell me. Okay, Keynesian Beauty Contest describes a beauty contest where judges are rewarded for selecting the most popular faces among all judges rather than those they may personally find the most attractive. This idea is often applied in financial markets whereby investors could profit more by buying whichever stocks they think other investors will buy rather than the stocks that have fundamentally the best value because when other people buy a stock, they bid the price up

allowing an earlier investor to cash out with a profit, regardless of whether the price increases are supported by its fundamentals. So it's basically this world in which we are trying to predict what other people will predict

as opposed to just predict what's going to happen in the real world. And that's where when we get to what do you think other people would look for in a partner, that's what you're doing. You're not just trying to think, what do I think they want in a partner? It's what do I think that they will say? And then that gets in social desirability. That gets a lot of this sort of second, third order thinking in it.

Yeah, that's really interesting. And so you could imagine that somebody who's paying more attention to revealed preferences might be able to get a leg up in that kind of way. That's an interesting idea. Yes. What were some of the biggest unanswered questions that you had after the study? Their implications, what it means for what people want at the moment?

So one of the questions that remains unanswered that I'm very interested in is, you know, in this study,

I'm a participant and I'm rating this partner on these traits. Now that's very useful. I mean, again, I'm a psychologist, so I believe in subjectivity is really important. And that if I want to understand my experience of why I like this person rather than that person, it helps to have me rate those traits. That tells me a lot. And, and, and we get the revealed preferences from those kinds of ratings. But if you were say a matchmaker, if you were an online dating company, you

You wouldn't have data like that if what you were trying to do was predict who was going to like whom. You would need people's self-reports on both sides, right? Or, I mean, you know, if we're just dreaming here, you might have, like, you know, independent coders, you know, rate people on their traits, okay? So what would we see for both revealed preferences and also that matching phenomenon is,

If we're using both people's self reports, right? So your self reports of your traits and then my ideals and trying to match that way. Now it is a basic rule with this kind of data that when you move from my judgment of the trait to your judgment of the trait and you try to predict something, that predictive relationship is going to go down. But

We have some wiggle room here. It might go down, but might still be useful. It might be the kind of thing that a matchmaker could use to predict who you're going to like more or less. It could also go down to basically zero. So that we don't really know yet. There are practical ways that these data could prove useful, but

Or we kind of end up in the same place where right now anybody who tells you that they have a matchmaking algorithm is probably just trying to sell you a secret sauce. Yeah, I saw you tweet an article, I think, from The Guardian that was quite critical about the effectiveness of online dating for finding soulmates. Yeah. A little bit of an assessment. People's sort of ambient dissatisfaction with this stuff.

Do you think that there is a way that your big data or big data like it could be folded into a dating service to make the matchmaking more accurate? I think it's possible. And that's what we need to see is if you can get a sample that's this large and

and you have people on both sides before they actually meet each other, maybe there is something you can do to predict good matches. Now, I want to be clear about what I mean by good matches, because if you have people's self-reports ahead of time, there are a few things that are very easy to do. It's very easy to predict who's popular. That isn't challenging. If somebody tells you, I tend to be popular with members of the gender that I'm interested in, guess what? They will be. So those kinds of

of self-reports tend to have accurate insight. What is much, much harder and we've never been able to do, and I've never seen anybody else who's able to do it, is to actually create that matching, that specific matching component or what I usually call compatibility. That it's not about your popularity.

It's really about the two of us fit together well. That's the sort of the holy grail in this space. Yeah, it's very interesting. I know that you've done a good amount of work about compatibility and sort of how people evaluate mate, mate evaluation theory and stuff like that. Yeah, yeah. I guess the one question that I have, or the sort of main potential flaw that I see in the study is that

When people have been doing their revealed preferences, that is still mediated by their own biases, right? They're describing either a real or imagined partner. And in that, it has to go through the filtering, which means that all of the muck of their...

cognitive makeup and their desirability and so on and so forth. How do we not know that they're maneuvering and manipulating the revealed person avatar through their own psychology? Oh, yeah, it's a great point. And, you know, that motivated reasoning

is in many ways, again, this is like one of the essential truths of close relationships, is that people are very motivated to see their partners in a positive light. There are some experiments, like 30 years old now, but there are really great studies where you tell people things like,

Hey, you know, good relationships have a lot of conflict. And then people will be like, Oh, yeah. Oh, my, my relationship has a lot of conflict. Yeah, this is not usually something that people would be willing to say. But when you tell them that actually would be a good thing. Oh, now they see the conflict in their relationships. Right now they're now they're willing to identify those moments.

People do that with traits too, right? So for a lot of these traits, there's a good version and a bad version of the thing, right? We were kind of talking about this earlier, but there's a good way of being sensitive, right? You're aware and seem to care that I'm feeling off today and you want to talk to me about it. But there's another way of being sensitive. It's like, oh man, this person is really touchy. And so what people will do is that when they're in a relationship that they're happy with,

They'll think somebody is the good version rather than the bad version of an attribute like sensitive. And so whenever we're getting people's own judgments, it's always going to have that sort of mucky self-report stuff in it. And that's not that useful if your goal is to try to predict who's going to like whom before they have a chance to meet each other and engage in that motivated reasoning one way or the other.

There's a third stage that I would love to look at. So, um,

One of the bits of research that I've done with David Buss was about the difference between what people will click on versus who they will click with. And the fact that algorithms seem to have a very good predictive power of being able to get you to swipe right on somebody. But when it actually comes to long-term compatibility, the predictive power is essentially zero. Yeah. Right, which presumably you've seen too. That makes sense, yeah. So we have, the way that I'm kind of conceptualizing is we have three. We have stated preferences.

Then we have revealed preferences. And then we have effective preferences for the long term as well. So I would love to work out, you said that you like this thing.

You ended up liking this thing. And what were the traits that ended up being effective over the long term as well? And that would require it to be, I don't know, longitudinal or for you to look at changes or like, what do you wish that you could change in a partner if you could? But that would be so cool because that would actually show not only what do people say they want, what are people attracted to? And then what is it that's got the best predictive power for effectiveness long term? Yeah.

Yeah, that's a great point. And, and one of the challenging things is that, you know, with any, the way humans go about initiating relationships is that you got to go through this stage to process, right? So I got to be sufficiently appealing to somebody initially and a first impression and a second impression and,

And then sort of get to those later stages. I think one of the challenges of modern dating is that as we have expected various online forms to be able to do the job for us,

it makes it a lot harder for some people who don't immediately convey a positive first impression to get to the later points where their other attributes or just the opportunity for some compatibility to grow and emerge, they can't get to that stage. Well, this is the emergence of the black pill, right? Yeah, right, right, right, right. But I think what the black pill misses is,

Um, and I suppose we can pin this on the red pill too, a little bit is that the answers in that space, I mean, correct me if you disagree, but the answers in that space are mostly about boosting my attributes so that I don't get cut off at the early stage of the process. Right. So that I'm appealing enough and look, lifting weights is great. People should, people should work out. People should take care of themselves, but

I think that what gets missed is that the thing that connects people and that gives them many opportunities is social connections, social networks, spending time with people. Spend time with people that you aren't going to hit on just for the sake of being around other people and being in those networks that grow and morph and change.

And, you know, I worry that we have like forgotten that. Yeah. There's certainly a sort of treating dating like a Petri dish. And, you know, humans are very bad at working out exponentials. They're very bad at working out compounding. But,

exponentials occur in social networks as well, right? You know, if you start to add one person into your friend group, into two people, into four people, into, you know, before you know it, the number of different connections between everyone and then all of their connections outside of it are very difficult to predict. I think this is, you know, I texted William Costello about this not long ago saying that I felt like my...

tumbling down the rabbit hole of evolutionary psychology has been so great at really sort of helping me to understand why humans are the way they are. But when it comes to the mating research, it's one thing that's fundamentally missed. It's missed by the red pill. It's missed by the black pill. It's missed by everybody. And it's even missed by evolutionary psychology, human behavioral ecology. Everybody misses it.

which is the phenomenological sense of falling in love with another person because it can't appear on a spreadsheet. There is no way that we can describe it, that we can measure it. It's this sort of

of weird sense of fluttering of butterflies and all the rest of it. And we can talk about she's trading her fecundity for his resources and her youth for the mate value. And in some form or another, red pill, black pill, evolutionary psychology, human behavioral ecology, the full work, social psychology, everyone will come up with their own way to describe this.

And no one is talking about, yeah, but sometimes you just sort of fall in love with somebody and you have no idea why and what the fuck's going on there. And I think that's really the X factor. It really sort of throws a wrench into the works of...

all of our presumed godlike predictive powers, it goes, yeah, that's out the window, sorry. Right. I think it is extremely chaotic, right? In the chaos theory sense of the term, right? That it really hinges on a set of interactions that

that can go in any of a wide variety of different directions. And, and once you fall in love with somebody, you can pick out. And I believe people when they say this, when they say, I fell in love with you because, um, because you're smart. And that time you were really interested in what had gone wrong with my job. And because my dog seems to love you.

And I believe people when they tell me that, that is why. And there was no way that we were going to flag those three things before you actually got together. There was no way to know. And that, and I think it's, you know, I find it both inspiring, but also kind of daunting. It's disenchanting as a researcher. Yeah, right. It's like, well, then what the hell are we going to do? What am I doing with my job? Yeah, but I will say,

We're not alone in this space. Personality researchers are getting into this stuff. People are starting to take this idiosyncratic personality. The things that describe you can't be captured by scales that you just feed to everybody. We're getting there. I think things are going to look different in 10, 20 years and are going to look pretty creative. Yeah.

You gave me an idea before when you were talking about the challenge that people have of getting past the first door, right? Of sort of getting someone to swipe right in them on a dating app or getting a phone number or in a bar or, you know, just getting a second sentence out of somebody if they're trying to speak to them. Yeah.

I think that there's a lot of sympathy that gets given to people who don't even get a chance. And that kind of makes sense. It's like an archetype of the poor down on his luck guy that would make a great partner, but isn't able to do it or whatever. But no one really ever talks about the reverse. The person who might be really phenomenal at making a sparky conversation, but just have like a completely unnerving,

objectionable, like, undeal-with-able personality, right? Or the, you know, the emotional stability, for instance. I'm going to guess if somebody is very low in emotional stability, they can probably kid many people. There may be lots of people listening that have been in relationships with people like this who have got past the front door, got into a relationship, everything seemed fine, and they go, oh, there's like a Jekyll and Hyde, Batman, and Bruce Wayne scenario going on here. And I...

Totally get it. People can't get a date. Sympathy is needed. How can we get these dudes to be more attractive? How can we get these women to be more interesting when they're talking to people? But there is a whole other cohort of people that need mate retention tactics as opposed to sort of mate attraction tactics. Yeah, right.

So there's a couple of interesting things here. First of all, I think the, the traits that you're describing that are probably most likely to show that pattern would be some of the more, you know, like narcissism and marquee Machiavellianism, like dark, dark triad stuff. I think actually the evidence on those traits being appealing at first is like a little mixed. I think they're like a little bit desired, but they certainly don't bode well for,

um, for people's longterm relationships and men and women. Yeah. And that's men and women. Yeah. So those, those, uh, those, uh, those attributes aren't great, but what's so interesting with things like emotional instability, it's like, the thing is like those attributes, you tend to come off badly at first and they're, they're not great in the long run. But what starts to happen is that, again, this gets into this like inclusion of the other and the self kind of thing is that, um,

Somebody who is anxious, if they get in a relationship with somebody who helps bring their anxiety down, maybe because they're super cool or just they make that person feel safe and comfortable, you can get a person who's like, look, I am still an anxious person, but I'm not anxious when I'm with my partner. I'm not anxious when I'm with these two friends. It doesn't even have to be romantic.

And I think it's not like a route to personality change necessarily, but it is a route to arriving at a place where you can have a happy, fulfilling relationship because you've found a way to tone down those traits, at least within the context of that relationship. How much do you think that the results you found both from a

and revealed preferences standpoint, how much do you think that they were always this way? How much do you think that these preferences have been subject to change across time? And can you think of any stated or revealed preferences that might be more or less subject to change? It's so interesting because I think about this all the time with the earning potential, you know, differences or lack thereof in particular. You know, I...

you know i watch a lot of movies and um i recently was watching the 1950s version of a star is born um and in that version you know the the the main female character becomes a major success um

And the man just absolutely falls apart, right? He cannot handle his partner being successful. And it is very understood in that movie, like very few men would be okay with this. Like this is absolutely emasculating that this is happening to him. So there is part of me that thinks, boy, this has got to be a recent phenomenon, right? Where men and women have the same revealed preferences for things like earning potential.

And then I read, you know, like Jane Austen and, you know, they're like, everybody's gold digging. So I,

I guess this is a long winded way of saying, I don't know. I suspect there are cultural trends that push these things around, but you know, I wish I had the data for what people, what people really wanted in, uh, in the 1800s or, you know, going back even further. But I, I, my guess would be is that as long as there was variability,

As long as there were rich women and poor women in the social circle, as long as there were rich men and poor men, that both men and women were going to gravitate toward the good stuff regardless of which gender they were pursuing. But that's just a guess. Yeah, it's an interesting one. I wonder whether...

increased globalization and increased inequality uh in that we can now see there's people who have obscene amounts of wealth that haven't that aren't like godlike creatures you know that isn't bestowed back down or blessed he's not the king he's not he's not somebody that's untouchable it's somebody that you can track their journey you know walter isaacson read 800 pages and you know elon musk one of the top richest guys in the world you know his story so

I wonder, you know, Candice Blake did some really fantastic work about how the proliferation of sexy selfies in areas of high income inequality. So basically it seems like women self-objectify more where they can see both how high they could climb with the right partner and how low they may be able to fall, essentially. So I wonder whether...

in a world that is basically that tuned up to 11, everybody can see how high they can climb. Everybody could see how low they could fall. Um, but then you throw the spanner in the works now, which is female learning potential and female financial independence. You know, women out earn men in their twenties. Uh, it's basically a motherhood tax. The pay gap is essentially just a motherhood tax now. And title nine has been reversed to women for every one man completing a four year us college degree. You know, um,

from the metrics perspective from 50 years ago, if you'd crossed off the M and the F and not shown people what it was, they would have said, oh, those are the guys and those are the girls. And you go, no, it's actually the other way around. Right, right, right, right. So yeah, you think...

What does it mean that ancestrally it should be the case that women would be more sensitive to the resource and status capacity of their partner in a world where they are now potentially out-earning their partner, especially during the time when they're looking to find a mate, etc.? You know, as you grow up a little bit older, men's desire to be obsessive and conquer and do mastery and stuff like that, plus motherhood tax, I think,

results in men on average earning and still is going to result in them earning more across their lifespan. It's a real sort of upside down. Well, how much can we change our, uh, both stated and revealed preferences? How much are we able to step in and consciously go, Oh, turn this down or turn this up? Yeah. It's a great question. You know, we've tried to do statistically,

some amount of experimental manipulations of people's stated preferences. It turns out it's pretty hard. It's pretty hard to change in a deliberate way people's ideas about what they want. We can do it,

with, you know, sort of these, it's like these conditioning paradigms, right? Where that you, you get people to experience positive outcomes with particular attributes and you do it more or less in various conditions. But all of these effects that we get, they're useful for testing things in the lab. But, you know, we're not changing whether people

say they care about earning prospects in an enduring way. But I do think it's important. Like when I think about the problems when it comes to, you know, gender relations today, I think a lot of it because we see these revealed preferences aren't so different.

I think a lot of it is that like our ideas and our expectations and those things can matter. I mean, we can get really mad when our, when we feel like our expectations aren't going to be met or, you know, when we think the world should be working this way, but it's working some other way. So I'm encouraged at one level that, well, it seems like if we want a route for change here, we just got to change, you know, the way people think about themselves, the way people see themselves and,

At the same time, I also know that any kind of lasting intervention is always a challenge. So what have you learned about where positive feelings about our partners come from? Like, what does it mean to say that we have a positive relationship?

sort of disposition toward our partners? What is that sense and where does it come from? - I mean, it's a deep question because in some ways it's the thing we focus on the most. I mean, again, I'm a close relationships researcher first and foremost, and the main thing we study is how positively people feel about their partners. And we think it's important because it's gonna predict

breakup and divorce. It's going to predict the health outcomes, right? But how exactly do you come to look at a rating scale and say, I'm at the top of this scale or I'm kind of middling? I think a lot of times we think it comes from

this general sense that you kind of lock in and you retrieve it the way you would retrieve the response to any other question like, oh, I think I'm extroverted.

But I also think that there are certain key major moments that happen for people in their relationships. Sometimes they're called turning points, right? And sometimes it can be a small thing like the one time that, you know, you made breakfast for yourself and not me. And that can be a real turning point. And I realized like, wait, do you even like actually care that I'm here?

Um, it can also be a very positive thing. Like the time that you skipped hanging out with your friends cause I'd had a bad day. And those moments too can also sort of push us in dramatic ways that, that again are kind of random. Like that same event could have happened to 10 other people and not have the same positive effect that it did on me.

Um, so I think it's some combination of those two things, right? Like overall senses that we kind of lock in and retrieve easily, but also these little moments that, uh, that, that end up being very memorable. And, and other than that, I don't really have a good sense of how people integrate those things. Do you think that it would be useful for people to become better predictors of their own

revealed make preferences. I wonder whether in your big cohort, the people who's stated and revealed had the smallest distance between the two are more reliably successful in relationships because they're better able to expedite finding partners that are like the ones that they want to be like.

That's a really interesting idea. And we've, I don't even know if we've ever collected the data that would really give us the right way to look at it. We at some point had some data where we tried to get a sense of who knew, like who was going to get an accurate sense of like how popular they are. And there are a few speed dating studies to that effect.

But not about like, oh, this actually –

someone's ability to stated and revealed are accurate they seem to be a little bit higher i mean if it was lower uh that would just throw a yeah in my hypothesis but um yeah that would be cool to find out yeah yeah that's a really interesting idea and the the general idea of looking at um

how satisfied you are with a whole suite of relationships. You know, it, this is where like polyamory research really has a, a leg up on what basic close relationship researchers do is because they can, in principle, see how you are reacting across multiple partners and,

And that has the potential for all kinds of fascinating insights because it suggests, you know, we might be able to see things like, you know, I say that I really care about attractiveness, but it turns out that I'm the kind of person who's I'm happier with the partner that I think is less attractive. Right. We would be able to see those kinds of differences in the way you are reacting to multiple partners at the same time. So unfortunately, those samples are pretty hard to collect, but they can reveal a ton of insights.

Paul Eastwick, ladies and gentlemen. Paul, I love this research. I'm fascinated to see what you do next. Where should people go if they want to keep up to date with all of the stuff that you're doing? That's a great question. You can follow me on Twitter, at Paul Eastwick. And yeah, I'm going to have a book coming out in about a year and a half. It's going to be a little ways off. So yeah, you can look for me then. But for right now, Twitter's good. I'd love to bring you back on when the book's out. That'd be great. That'd be great.

Hell yeah. We did it, man. Appreciate it. Thanks so much.