Under the Influence (with Jo Piazza) - April 27, 2025


Sunday Nice Things： Normal Curves

Topics
sexy body odor sending spit samples for dating sweaty t-shirt study sweaty t-shirt study bar charts and numerical data correlated observations data shenanigans
Will Your Kid Have to be a Brand to Survive？ We Survived Girls Gone Wild Just to Get Tradwives？

Episode Stats


Length

59 minutes

Words per minute

187.15

Word count

11,181

Sentence count

1,204

Harmful content

Misogyny

42

sentences flagged

Toxicity

15

sentences flagged

Hate speech

54

sentences flagged


Transcript

Transcript generated with Whisper (turbo).
Misogyny classifications generated with MilaNLProc/bert-base-uncased-ear-misogyny .
Toxicity classifications generated with s-nlp/roberta_toxicity_classifier .
Hate speech classifications generated with facebook/roberta-hate-speech-dynabench-r4-target .
Topics generated with Qwen2.5-3B-Instruct.
00:00:00.680 I don't think I can be your friend, Isabella said.
00:00:04.060 Rebecca's stomach flipped.
00:00:06.180 This is the love story of real hinge couple Isabella and Rebecca,
00:00:09.900 written and read by me, Temi Denton-Hurst.
00:00:12.360 Listen to the free audiobook now.
00:00:22.520 Hey, all.
00:00:24.120 Jo here, and I am dropping a new episode of a new podcast into your feed
00:00:28.560 for Sunday nice things. Normal Curves is a podcast about sexy science and serious statistics.
00:00:36.760 If you've ever tried to make sense of a scientific study and the numbers behind it that
00:00:41.000 just didn't seem to add up, you are not alone. I love this show so much. The hosts are two
00:00:48.700 professors who discuss academic papers, fun style, like funsies. There's no jargon,
00:00:54.860 It's irreverent.
00:00:55.760 It's PG-13 in a way that I love.
00:00:58.940 So get in there.
00:00:59.860 Get a listen.
00:01:00.440 You can binge it all at Apple, Spotify, wherever you get your podcasts.
00:01:03.940 Normal Curves.
00:01:05.060 You're going to love it.
00:01:05.820 It's a lot of fun.
00:01:12.900 I think you're going to like this, Tristan.
00:01:14.780 Really?
00:01:15.420 And by like, I mean hate.
00:01:17.940 So I might have some criticisms is what you're saying.
00:01:20.000 Maybe a couple.
00:01:24.600 Welcome to Normal Curves. This is a podcast for anyone who wants to learn about scientific studies and the statistics behind them.
00:01:32.440 It's like a journal club, except we pick topics that are fun, relevant, and sometimes a little spicy.
00:01:39.440 We evaluate the evidence, and we also give you the tools that you need to evaluate scientific studies on your own.
00:01:46.340 I'm Kristen Sinani. I'm a professor at Stanford University.
00:01:49.460 And I'm Regina Nuzzo. I'm a professor at Gallaudet University and part-time lecturer at Stanford University.
00:01:54.600 Stanford. We are not medical doctors. We are PhDs, so nothing in this podcast should be
00:01:59.340 construed as medical advice. Also, this podcast is separate from our day jobs at Stanford and
00:02:04.900 Gallaudet University. Kristen, today we are going to look at a classic paper in the scientific
00:02:10.000 literature, commonly known as the Sweaty T-Shirt Study. Interesting nickname. It is, and along the
00:02:17.000 way, we are also going to learn about things like pheromone dating parties, pheromone dating 0.96
00:02:22.100 websites and choosing your dating partner by sniffing up his armpits. 0.53
00:02:26.780 Okay, that sounds interesting.
00:02:28.460 It's a good paper, weird but fascinating, and a ton of studies have then taken that
00:02:34.420 and extrapolated even further.
00:02:36.680 All right, so the idea here is that we all might be choosing our mates based on how they
00:02:41.980 smell.
00:02:42.640 Am I getting that right?
00:02:43.520 Absolutely.
00:02:44.080 The claim is this, that women prefer the smell of men whose genetics are different from them
00:02:51.920 and we're going to take a closer look at that.
00:02:54.460 But we're also going to touch on a few topics
00:02:57.100 that people may have heard about in stats class,
00:03:00.040 things like bar charts, correlated observations,
00:03:03.520 but people are probably not learning in stats class about sexy body odor.
00:03:08.540 Well, correlated observations, that is something that I teach.
00:03:11.720 I have a lot to say about that. 0.53
00:03:13.560 But yeah, I don't teach it in the context of sexy body odor. 1.00
00:03:16.720 I use the example of eyes because you have two eyes,
00:03:19.980 So those eyes are correlated within a person, right?
00:03:22.760 That's kind of boring. 0.81
00:03:24.320 Armpits.
00:03:25.180 We've got two armpits.
00:03:26.260 Oh, well, I could switch examples, and that might wake a few students up.
00:03:29.980 Body odor. 0.98
00:03:30.920 It always works.
00:03:32.080 Yes, definitely.
00:03:33.940 But, Regina, I want to know, what goes on at one of these pheromone dating parties that you mentioned?
00:03:38.800 Did you ever go to one?
00:03:39.920 I have not.
00:03:41.080 Sadly, I have just read about these.
00:03:43.100 Okay.
00:03:43.500 Yeah.
00:03:44.120 So how does it work?
00:03:44.580 I can tell you the whole concept, though.
00:03:48.040 So these are patterned off of the sweaty T-shirt study.
00:03:51.060 Okay, the study that we're going to talk about upcoming.
00:03:53.020 Mm-hmm.
00:03:53.620 So here is the idea.
00:03:55.480 Men, women, both, you sleep in a clean T-shirt for two nights.
00:03:59.440 Okay. 0.68
00:03:59.840 Yeah, no deodorant, just like all you.
00:04:02.300 And then you bring it into a bar, party at a bar, in a sealed plastic bag.
00:04:06.880 So you're not allowed to wear any deodorant during those whole two days?
00:04:10.480 Yep.
00:04:11.540 Yeah, or at least not at night.
00:04:13.460 Oh, okay.
00:04:13.880 I don't think there were a lot of rules here.
00:04:16.740 Oh, okay.
00:04:17.020 Okay, I'm thinking of this like a scientist.
00:04:20.320 Let me put my scientist hat aside for now.
00:04:22.880 You sleep in your t-shirt at night, and it picks up your scent.
00:04:26.420 I did read about one person who let their dog sleep in their t-shirt,
00:04:31.040 and they brought that in.
00:04:33.760 So clearly, this is not a well-controlled study.
00:04:37.360 I want to know what happened with the dog t-shirt, though.
00:04:40.140 Like how many people?
00:04:41.320 I would probably smell the dog and say, yep, I want that one.
00:04:44.500 I love my dog.
00:04:46.080 I mean, yes.
00:04:47.640 Yes, that was crazy.
00:04:48.560 Okay, yes.
00:04:49.180 So uncontrolled study, uncontrolled experiment.
00:04:52.180 They recommend that you sleep in it for two nights
00:04:55.000 and just to kind of infuse it with your own body odor.
00:04:57.960 You're bringing the T-shirt in.
00:04:59.640 You're allowed to shower before the party.
00:05:01.140 You are allowed to shower.
00:05:02.860 Probably encouraged to shower before the party.
00:05:04.400 Okay, yes. 0.80
00:05:04.640 So you bring this awful gross T-shirt in.
00:05:07.240 There we go.
00:05:07.760 You bring it in a steel plastic bag.
00:05:09.180 Oh, nice.
00:05:09.740 That's very hygienic.
00:05:10.760 The organizers take it and they give it an anonymous label and gender.
00:05:16.080 Oh, it is kind of like a little science experiment then.
00:05:18.180 It is kind of, although I think we could probably improve on it.
00:05:21.800 But they throw it all, all the bags on a big table in the middle of the bar.
00:05:26.000 Okay, you're at a bar.
00:05:26.840 You're at a bar, first of all.
00:05:28.400 Right.
00:05:28.680 I mean, what other way are you going to smell the T-shirt?
00:05:32.540 You need a few drinks before you can do this.
00:05:34.260 You need a few drinks before you do this.
00:05:35.660 So you're wandering around, you're having drinks, you're chatting people up,
00:05:38.820 and at the same time, you're sticking your head in these plastic bags.
00:05:42.780 Oh, you go over and smell.
00:05:43.200 And you're smelling the sweaty T-shirts.
00:05:45.000 And if you find one you like, then what happens?
00:05:48.400 Then what happens?
00:05:49.120 Then you walk with the bag up to the organizer's table up front,
00:05:53.200 and they take a photo of you holding the bag up.
00:05:56.320 Ooh.
00:05:56.780 And then they project that up onto a big screen so everyone can see.
00:06:00.720 This is not a party for introverts.
00:06:03.000 Definitely not.
00:06:05.640 And, yeah.
00:06:06.720 Okay, and so you're holding the bag with the number,
00:06:08.860 so the person who owns that bag that they can see,
00:06:14.240 And are they getting to smell your bag then?
00:06:16.400 They are not.
00:06:18.000 So it's not double blind.
00:06:19.120 Oh, goodness.
00:06:20.060 I know.
00:06:20.380 See, I told you.
00:06:21.460 We can improve on the state of the gun.
00:06:22.340 This is not a good science experiment.
00:06:23.640 Okay.
00:06:23.940 But they're just, if they like your look.
00:06:26.160 If they like your looks.
00:06:27.140 Then they're going to go find you.
00:06:28.700 So what if your picture gets projected and then nobody comes and finds you?
00:06:33.100 I know.
00:06:33.240 I don't know.
00:06:33.580 I don't know.
00:06:34.580 That would be crushing.
00:06:37.140 Okay.
00:06:37.400 But this is in the background.
00:06:38.460 We had a lot of alcohol.
00:06:39.420 So maybe you don't care.
00:06:40.380 So maybe.
00:06:41.260 Right.
00:06:41.480 Okay.
00:06:41.700 Maybe you just did not even notice.
00:06:43.140 So at least the person who picked the bag, they are basing their choice without seeing the person, without knowing their salary, right, just on this primal instinct.
00:06:52.800 Well, maybe we should all be picking that way.
00:06:55.060 Who knows, right?
00:06:55.880 Is this better? 1.00
00:06:56.560 This is actually the epiphany that the woman had who started a lot of these parties. 0.88
00:07:01.360 Oh, okay. 1.00
00:07:01.960 So she started a business where she was doing this.
00:07:04.020 So she said that she had always been dating around.
00:07:07.440 She was in her 20s dating based on, you know, salary and resume and looks.
00:07:11.660 Until one time she dated someone she just had an animal attraction to.
00:07:16.040 And she said it went so much better.
00:07:18.640 Interesting.
00:07:19.620 And she was obsessed with the way he smelled.
00:07:22.580 Oh, interesting.
00:07:23.520 And she knew about the Sweaty T-Shirt Study?
00:07:25.480 She had read about the Sweaty T-Shirt Study.
00:07:27.620 So she set up these parties to share her epiphany with the entire world.
00:07:31.560 That's where the organizers are from.
00:07:33.540 She provided the organizers.
00:07:34.920 Oh, interesting.
00:07:35.840 In these different cities, London, New York, L.A.
00:07:39.060 So was it successful?
00:07:41.180 I mean, did she match a bunch of people?
00:07:43.580 And are they married and living happily ever after now?
00:07:46.480 I don't think they had a longitudinal follow-up.
00:07:48.500 They had outcomes.
00:07:50.100 What was the primary outcome?
00:07:51.480 What was the primary outcome?
00:07:53.200 Marriage, right?
00:07:54.220 Or maybe just hooking up for the night.
00:07:56.480 Yes, we don't know.
00:07:57.940 Sadly, the world will never know unless we start one up again.
00:08:01.480 But that is the idea behind the pheromone.
00:08:04.920 And this was based, again, on that Sweaty T-shirt study, which we're going to get to.
00:08:08.400 I'm looking forward to talking about the study itself.
00:08:10.140 But before we do that, I think it's probably easiest if we talk about the pheromone dating sites that I alluded to earlier.
00:08:18.500 Oh, right. Okay.
00:08:19.020 Because there's more science there.
00:08:19.680 So another business model off of the T-shirt study.
00:08:22.080 Wow.
00:08:22.560 There is another one.
00:08:24.060 This one has more science, so you're going to like it.
00:08:26.120 So these pheromone dating sites, I wrote about one, actually, back in 2008.
00:08:31.160 Oh.
00:08:31.440 This is when you were writing the science of sex column for the L.A. Times.
00:08:36.300 The mating game, the science behind mating, dating, and sex.
00:08:40.160 Yes.
00:08:40.800 Not Regina's sex life.
00:08:42.020 I always feel like I need to point that out.
00:08:43.260 Yes.
00:08:43.880 So I dug up my old L.A. Times story.
00:08:46.040 Ooh, I want to hear.
00:08:47.340 Here's just a quick quote.
00:08:49.540 Okay. 0.92
00:08:49.920 The dating site, quote, offers to find you a lover who smells good.
00:08:54.140 If all goes well, you'll get a lusty good smell,
00:08:57.180 the kind that makes you bury your face in your mate's pillow the next morning to catch the
00:09:03.080 lingering site. Well, I love your writing, Regina. That's very evocative. Also promising a lot.
00:09:08.800 And question for you, are you sending a sweaty t-shirt to the dating site and then they're like
00:09:13.920 cutting it into little pieces and sending it to a potential man? Like, how does this work?
00:09:18.320 Logistics are very important here. This is the cool thing, though. There's no smell. It's all
00:09:23.780 genetics. Okay, whoa, wait, what? No actual smell. All of this thing is just genetics. You send in
00:09:29.280 a sample of your spit, and they use a few particular genes from your DNA to suggest
00:09:36.900 your matches. Okay, so they're using genes that they think are related to pheromones that people
00:09:41.720 might be smelling. Okay. It's like, instead of the DIYing, it's by going around snipping up
00:09:50.140 people in a bar which could be confounded by like the alcohol you're drinking and things yes
00:09:54.740 right this is like cutting straight to the chase and saying okay there's genetics behind these
00:09:59.640 pheromones and are they matching you like sight unseen you don't you never get to see a picture
00:10:04.920 or know the age or anything of the person this site does not exist but from what i can understand
00:10:09.760 is they just suggested matches and then it went on like a traditional thing so you can still say
00:10:14.160 you could still scream i want one where they're just like nope this is the person you compatible
00:10:18.380 with go meet them and that's it done that that would actually make a great reality show that
00:10:23.000 would make a great reality show regina we're on to something yeah all right but you said that the
00:10:28.660 dating site no longer exists so it went defunct it went defunct but seriously if we started up as
00:10:34.160 a reality show love it for twiff and love it oh my god we shouldn't be giving this away
00:10:39.180 this idea we should be keeping this in a secret box somewhere we just trademarked it
00:10:43.300 this is the idea so you also got a lot of perks like you said they were promising a lot so they
00:10:52.000 were not just promising you someone who like who smell right so it's not just that you are going
00:10:58.460 to get somebody you like the smell of you're also going to get somebody who what better sex 0.94
00:11:02.920 oh increased fertility oh healthier kids less cheating and more orgasms for women only 0.99
00:11:11.900 But that's okay. 0.82
00:11:13.480 The men appreciate that anyway.
00:11:15.180 Right, yes, yeah.
00:11:16.780 Wow, so they're saying if you pick your mate this way,
00:11:19.820 then it really gets down to all the animalistic thermos,
00:11:23.920 all about the sex and mating reproduction.
00:11:26.580 That will all go well.
00:11:27.380 Now, he might not have any money.
00:11:29.940 He might be on the street, but the sex will be good. 0.93
00:11:32.240 But the sex would be good and you'd be happy.
00:11:34.420 I'd love to know if there were success stories from this,
00:11:37.520 but it's defunct, so maybe it didn't work
00:11:40.120 Or maybe this just wasn't a good business model.
00:11:42.120 I'm wondering if people didn't like sending in their spit.
00:11:44.800 And then, because then when you tell people how you met, it was like, we met through drool.
00:11:49.780 That does sound kind of gross.
00:11:51.140 Like, maybe it's just not so sexy.
00:11:53.040 All right.
00:11:53.280 So, Regina, this company was matching people based on genetics, but how were they matching them?
00:11:58.560 How was that?
00:11:59.080 What were they looking for in genetics?
00:12:00.420 Yeah, yeah.
00:12:00.860 So, it's interesting.
00:12:01.900 They were looking to see if you and your potential mate had different genes, just a few genes in one particular area.
00:12:09.900 of a chromosome. And that area is called MHC, major histocompatibility complex. You might have
00:12:17.080 also heard this called HLA, human leukocyte antigen. Yes. And those are different names for
00:12:22.600 the same genes, actually. And these are immune genes. They have something to do with the immune
00:12:26.760 system. Immune system. And Regina, when you mentioned for today's podcast, we were going
00:12:31.200 to be talking about HLA genes. I remembered that I wrote an article back in 2011, where I talked
00:12:37.160 about HLA genes. You did? Yes. And the context was, I was writing, I think this is the only time
00:12:42.980 I've written about sex. I was writing about sex. I was writing about ancient sex though. Old people
00:12:48.380 having sex. Sorry, archaic humans. So sex during evolution of modern humans because modern humans 1.00
00:12:56.560 and archaic humans like Neanderthals were on the earth for a period of time at the same time. And 0.54
00:13:02.460 they actually intermated. They had sex with each other. And we know this because you can find 0.98
00:13:07.100 little pieces of Neanderthals like teeth and pinky fingers and people have sequenced the DNA from
00:13:13.520 those samples. And you can show that there are some genes that we inherited from Neanderthals
00:13:19.720 in the modern gene pool. Yes. And these are the HLA genes? Yes. I was writing about research from 0.59
00:13:25.180 Stanford and one of those genes that we inherited from Neanderthals is a specific HLA gene, which is
00:13:31.320 showing how important these HLA genes are because we kept this gene throughout evolution and it must
00:13:37.800 be benefiting us in some way, maybe because HLAs, it's good if they're diverse. I am so jealous that
00:13:43.320 you wrote about Neanderthal sex because now I want to. That is on my bucket list. And because the 0.99
00:13:50.340 gene I was talking about was an HLA gene, I had to give a little context at the beginning of the
00:13:54.200 article. So I said what HLAs are and I'm going to read you one of the sentences from my article.
00:13:59.340 HLA diversity is so important that it may even influence mate selection.
00:14:06.240 Studies show that people are attracted to the sense of prospective sexual partners with disparate HLA types.
00:14:13.720 I think I wrote about the study t-shirt study.
00:14:16.900 That is, oh, absolutely.
00:14:18.140 Right, I mean, it's exactly what we're talking about today.
00:14:20.300 I had completely forgotten that I'd written that.
00:14:22.640 I didn't maybe do my due diligence as a journalist because I just said studies show.
00:14:27.040 Everyone knows.
00:14:27.880 I did not go back to the original source.
00:14:30.860 I must have read that on the internet, like there's a bunch of studies that show this,
00:14:34.600 and I just took that at face value.
00:14:36.460 So maybe we're doing a little penance today.
00:14:38.560 Maybe I could have fact-checked that better.
00:14:40.980 You know, it's fascinating that you brought this up,
00:14:43.420 because people do reference the sweaty t-shirt study all the time.
00:14:47.200 It's everywhere.
00:14:47.960 I referenced it without even knowing that that's what I was referencing.
00:14:50.340 But that's great that we're going to actually go back and, you know, 13, 14 years later here,
00:14:54.740 I'm going to actually learn what was the science behind this.
00:14:57.080 What is the science, right? Or is there any science behind it?
00:15:00.160 The other thing I wrote about HLAs in the background, just to give people a sense of what genes we're talking about,
00:15:05.380 this is probably the context where most people know of HLAs, and that's organ transplantation.
00:15:10.060 Because when we talk about matching people for organs, what we're really matching is those HLA genes.
00:15:16.160 Be the match.
00:15:17.540 Yes, exactly.
00:15:18.420 It's because these genes, to get into the science behind why that's important,
00:15:23.080 The genes control how the immune system is recognizing those things floating around our body,
00:15:28.160 what is supposed to be there, what is self, versus what is not self, not supposed to be there.
00:15:33.620 Exactly.
00:15:34.140 Invaders, bacteria, viruses, fungi, parasites.
00:15:37.600 Right.
00:15:37.860 All kinds of, or someone else's organ.
00:15:39.920 Right. Normally, you want your immune system to react to those things, but in the case of
00:15:43.560 organ transplant, you don't want the body to see that as foreign and to reject it.
00:15:47.760 So you want to be as similar as possible to the organ donor in terms of your HLA genes, yeah.
00:15:53.620 It's hard to find an exact match, like a perfect match, yeah, if you're not related to them.
00:15:59.180 I think in the general population, finding an exact match, it's like one in 100,000.
00:16:04.580 Right. It's quite rare, yeah. That's why often family members are the best donors.
00:16:08.280 So with dating, though, we want kind of the opposite.
00:16:12.060 We want someone with a different gene.
00:16:14.460 Right.
00:16:14.860 We don't want somebody similar because you don't want to be marrying somebody in your family.
00:16:19.620 Unlike for organ transplantation, you're looking for as diverse as possible.
00:16:24.300 Genetic diversity is good in evolution.
00:16:26.980 Hybrid vigor.
00:16:27.800 Hybrid vigor.
00:16:28.800 So this is getting to like the immune system.
00:16:31.220 You want to give your kids the best possible immune system. 0.70
00:16:34.320 Right.
00:16:34.600 And it's kind of interesting how the genes do it.
00:16:37.660 Like, why do you want someone with different genes?
00:16:39.780 I had to go look this up.
00:16:41.120 So the genes, there are these six major HLA genes.
00:16:44.340 And my favorite is HLA-DR.
00:16:46.340 This gene comes in like 11,000 different variations or flavors.
00:16:51.160 You're talking about alleles.
00:16:52.160 Alleles.
00:16:52.460 Right, yes.
00:16:53.200 I like to think of them as ice cream flavors.
00:16:55.240 Yes, slightly different versions of the gene.
00:16:58.600 And the HLA genes have the most diversity of any genes in the human body.
00:17:04.300 Highly polymorphic.
00:17:05.220 Right.
00:17:05.840 I like using the technical term, you know, if you compare that to say an eye color gene or something,
00:17:10.060 There's probably very few versions of that.
00:17:11.900 But the fact that we have so much diversity in these HLA genes tells you that diversity in those genes must have been important in evolution.
00:17:18.240 Important, which is why we're still hanging on to the Neanderthal genes there.
00:17:22.820 And so it works because, okay, you've got these 11,000 different alleles, you know, flavors.
00:17:28.020 They each specialize in recognizing a slightly different foreign invader, which is just mind-blowing.
00:17:34.140 So there are some that are really good at detecting the bacteria that causes leprosy.
00:17:41.100 Oh, interesting.
00:17:41.780 And another one, hepatitis C.
00:17:44.020 And another one, HIV, actually recognizing the HIV virus.
00:17:48.440 Right.
00:17:48.600 So if you carry that particular allele, you might be less likely to get HIV even if you were exposed.
00:17:53.640 Mm-hmm.
00:17:53.840 Interesting.
00:17:54.780 So you can see why you want your partner to have different allele, not the same.
00:17:59.300 That makes total sense.
00:18:00.380 Kind of like an investment portfolio.
00:18:02.040 Right.
00:18:02.900 Diversify, diversify. 1.00
00:18:03.980 Okay, Regina, so we want to choose a mate that has very different HLA genes from us.
00:18:09.940 But if I were guessing, I would say like, okay, when you're looking for a mate, often we, you know, opposites attract.
00:18:16.520 We might be looking for somebody who looks very different from ourselves.
00:18:19.600 Exotic.
00:18:20.400 I definitely went for the exotic types when I was dating. 0.99
00:18:22.940 The exotic genes.
00:18:23.400 And I thought maybe I'm doing that because, you know, hybrid vigor.
00:18:26.880 So you're saying, though, it's not just looks.
00:18:29.800 You're saying that we might be able to actually smell each other's HLA genes.
00:18:33.100 It's like, how does that work?
00:18:34.120 Smell.
00:18:34.720 It is fascinating, especially because researchers are not sure how we smell each other's genes.
00:18:40.320 It sounds a little far-fetched.
00:18:42.360 Is there anything to base this on?
00:18:44.160 The idea that maybe these HLA genes are, you know, creating these chemicals and then it's coming out in our sweat.
00:18:50.040 What they do have evidence for is this whole thing in mice, that mice can smell each other's pee,
00:18:57.660 and they can smell the different genes in their pee
00:19:00.800 and that they're choosing to mate preferentially
00:19:04.120 with mice that have different immune system genes.
00:19:06.740 Oh, okay.
00:19:07.320 So there's some evidence in mice that this is true.
00:19:09.820 They saw this in animals
00:19:10.900 and no one had studied it in humans yet.
00:19:13.320 And that's where you get the sweaty t-shirt study.
00:19:15.780 That's the origin of the sweaty t-shirt study.
00:19:17.660 That is it right there,
00:19:18.920 which I think we are now ready to talk about.
00:19:21.200 I can't wait to hear about it.
00:19:22.380 you want to impress them on a first date but also play it cool so what do you do i'm rufy thorpe and
00:19:34.500 i wrote and read a real love story about a hinge couple that navigated exactly that
00:19:39.120 listen to the free audiobook now
00:19:41.340 today we're looking at the claim that women prefer the smell of men whose genetics are
00:19:50.060 dissimilar to them. And we're now going to talk about the study behind this claim, which is the
00:19:57.020 sweaty t-shirt study. And I'm really looking forward to hearing the details of this study.
00:20:00.720 Yeah, I think it's one of those studies that people have heard about. Maybe they remember
00:20:05.900 conclusions, but if you press them, they can't really tell you any details.
00:20:09.960 Guilty, wrote about, and didn't look up any of the details. Took at face value, yes.
00:20:16.160 Understandable, understandable.
00:20:17.500 Spoiler alert right here.
00:20:19.660 Let's just say it was not as rigorous as I was hoping it would be.
00:20:24.760 Let's just say that.
00:20:25.660 So I'm going to hit the highlights here.
00:20:27.420 Some places you're just going to have to trust me, Kristen.
00:20:30.960 Okay, we're not going to go over every bad detail yet.
00:20:33.120 We'll put more info in the show notes.
00:20:35.540 Dig in more.
00:20:36.520 Okay, because it had problems with how they designed the study,
00:20:40.360 the different ways they analyzed the data,
00:20:42.680 how they reported their methods, the results, the conclusions they made. It was a whole bit.
00:20:48.660 Oh, no. Well, I mean, look, first of all, I'm shocked that there was a bad study, right?
00:20:54.220 Those exist, right? I'm not surprised that the study was bad. I'm disappointed, though,
00:20:58.140 since this is widely cited. I know. It's not all horrible, but let's get into some of those
00:21:04.140 details. Sadly, the paper does not have the word sweaty or t-shirt in the title. I know. I was
00:21:10.960 hoping. It's called MHC Dependent Mate Preferences in Humans. Such a boring academic title.
00:21:17.400 Published in 1995. Lead author is a guy named Klaus Wedekind, an evolutionary biologist at the time.
00:21:23.420 He was a lecturer at the Zoological Institute at University of Bern in Switzerland. Okay,
00:21:27.940 Switzerland. So let's get a little bit into methodology. Right. Say design. First of all,
00:21:34.020 sample size. Yes. How big was this study? I want to know. 93 students. Okay, not the worst
00:21:38.760 sample size ever? That university, 49 women, 44 men. Okay. So remember those numbers. Yeah. They
00:21:44.560 were all about 25 years old on average. And remember from the study goals, there were two
00:21:49.860 things they needed to measure that we're going to talk about. First of all was what male body odors
00:21:55.220 the women preferred. Right. They also needed to look at the different HLA gene versions that the 1.00
00:22:01.240 men and the women had. Right. So I'm guessing that the preference for smells was through sweaty 0.57
00:22:06.040 t-shirts because it's called the sweaty t-shirt study so they brought in sweaty t-shirts like
00:22:09.620 the pheromone party men wore a new cotton t-shirt for two nights and unlike the pheromone party this
00:22:15.700 was actual science so they had a list of instructions no smoking no drinking no sex
00:22:22.180 so they didn't want to contaminate their natural smell that makes sense yeah animal smell and the
00:22:27.920 women didn't have to wear the t-shirts because this is just women smelling men right but the 1.00
00:22:31.100 women had a protocol to follow. Oh, okay. The women had to use a nasal spray for two weeks 1.00
00:22:35.740 before they experiment to keep their nose moist and virus-free. Right. So they want them to have 1.00
00:22:40.620 the full sense of smell and their smell to be protected prior to the study. Okay. Also given
00:22:44.840 a copy of a novel to read to sensitize their smell perception. A book. Okay. What's the novel?
00:22:51.360 It's called Perfume, the Story of a Murderer. Okay. From 1985. Wow. Have you read it? I have
00:22:58.340 not read it. No, have you? I actually downloaded it and started reading it. Oh, really? Okay.
00:23:03.620 Well, is it good? Oh my God. It's so weird. It is. It's about a guy in 18th century France who
00:23:10.940 is a super smeller. So we can smell things down to the molecule. Okay. He murders women to capture
00:23:16.060 their smell and make perfume from them. Well, I think I, I'm not sure I'd want to be in this
00:23:20.640 study after reading that, but what message is that sending to the women? Yeah. Was that just
00:23:25.140 It's supposed to prime that they're aware of their sense of smell.
00:23:27.900 They're aware of their sense of smell and how important it is.
00:23:31.520 So the t-shirts then, the guys had infused their lovely scent into the t-shirts for two nights.
00:23:37.280 Each woman got six t-shirts to smell.
00:23:39.320 Three were from men that were classified as genetically dissimilar, the exotic genes. 0.96
00:23:46.040 And three were from men who were classified as genetically similar.
00:23:49.000 And the women snipped each t-shirt.
00:23:52.020 Regina, how did they determine which men were genetically exotic or genetically similar?
00:23:57.560 What was the classification scheme?
00:23:59.320 Yeah, they were not transparent about that, actually.
00:24:02.200 I know, like, how much overlap in genes do you need before you're defined as genetically similar?
00:24:08.660 Did you have to be eligible to be, you know, an organ donor for them?
00:24:12.780 Or just kind of similar.
00:24:14.120 So they didn't give any details in the paper?
00:24:15.860 No, they did not.
00:24:16.700 And they were only able to look at three of the six HLA genes, so that's not great out there.
00:24:22.760 That's not great, yeah.
00:24:23.900 Yeah, when you make a judgment call like they're doing here, you really need to report it.
00:24:30.080 Transparency is key here.
00:24:31.960 Right, otherwise we can't replicate it.
00:24:33.860 Right.
00:24:34.300 We don't know if it's valid.
00:24:35.440 Right.
00:24:36.440 Another thing they weren't transparent about, Kristen, they did not report all their results, actually. 0.99
00:24:41.820 The women rated three things, how intense the body odor was, how pleasant it was, and how sexy it was.
00:24:49.920 And I'm guessing that we care the most about how sexy the smell was to the women, right? 0.91
00:24:55.780 Actually, they never reported the results on sexy.
00:24:58.900 What? Are you kidding?
00:24:59.720 They never reported them.
00:25:00.740 We don't get the data on that at all.
00:25:02.200 Only pleasant.
00:25:03.060 Okay, so this is making me very suspicious because—
00:25:08.380 I told you.
00:25:09.040 you usually are going to report all your variables,
00:25:11.920 especially since sexiness is the relevant one here.
00:25:14.600 Like, we're not going, oh, I'm going to date that guy because he's pleasant.
00:25:17.140 Because he's pleasant.
00:25:17.940 We do not.
00:25:18.400 We date men that are sexy.
00:25:21.560 Right, exactly.
00:25:22.360 So, okay.
00:25:23.340 And there's only three of them, so they could have reported all of them.
00:25:26.040 They could have.
00:25:26.580 It's not like it takes a lot of space in the paper.
00:25:28.880 But it makes me worried that something happened like the following.
00:25:32.620 I don't know this for sure, but this is where my hackles go up.
00:25:36.100 That maybe they analyze the data for pleasantness.
00:25:38.340 They analyzed the data for sexiness and pleasantness just made statistical significance and sexiness just missed statistical significance or something like that.
00:25:45.660 And they're like, oh, I don't want to we don't want to show the one that wasn't a positive result.
00:25:49.480 And another thing they were not transparent about, Kristen, they didn't actually report any numerical summaries of their data.
00:25:56.240 They used bar charts to show the average value on these ratings, like how pleasant the odor was.
00:26:02.940 But they never gave the exact numbers.
00:26:05.200 So you had to, what, take a ruler, draw a line across to the y-axis and guess or estimate those numbers?
00:26:11.880 That is basically what I had to do just to recreate the numbers for us to talk about here.
00:26:16.200 That's annoying.
00:26:16.940 Yeah.
00:26:17.840 And even worse than that, Kristen, can we do a little statistical detour here about bar charts and what they are actually supposed to be used for?
00:26:26.320 Sure.
00:26:26.720 Because it is not this.
00:26:28.280 This is an inappropriate use of bar charts.
00:26:31.680 Old man shakes fist at clouds.
00:26:34.160 That is me.
00:26:35.200 Yes, I know this is one of your pet peeves, Regina.
00:26:37.580 Can you explain why for our audience?
00:26:40.040 The data we have here are numbers.
00:26:42.520 They're ratings on a scale.
00:26:44.500 Right.
00:26:45.000 These are what we call numerical data.
00:26:47.040 They are numbers that you can add, subtract, multiply, or divide.
00:26:51.000 But bar charts are not meant for numerical data.
00:26:53.420 They are meant for categorical data.
00:26:54.800 They are not meant for numerical data.
00:26:57.240 Categorical data would be something that, believe it or not, has a category.
00:27:01.520 It's like how many people have blue eyes versus brown eyes.
00:27:05.200 green eyes. Now, bar charts make sense for categorical data because the height of the bar
00:27:10.940 corresponds to something, to how many people are in each category.
00:27:15.520 Right. Like this many people have blue eyes, this many people have green eyes.
00:27:18.880 Right. And you can almost even picture each of those people stacked up standing on top of each
00:27:23.980 other. Right. And that's the height of the bar. It has a physical correspondence.
00:27:28.420 Right. But the bar doesn't mean much when you're talking about numerical data. Typically,
00:27:32.540 people report the mean and they just make a bar that starts at zero and ends up at the mean,
00:27:37.060 which actually makes no sense. No sense. Because why is it starting at zero, right? The height and
00:27:42.320 zero don't correspond to anything physical. Like what if you had a mean that is negative,
00:27:49.120 right? A negative mean, would you have an upside down bar chart? I want to see an upside down
00:27:53.080 bar chart. Right. The only thing they're displaying here is the mean, but that's just
00:27:57.320 a single number. It's like we could erase the entire bar and just give a point. Exactly. So
00:28:02.240 Just put the number in a table.
00:28:04.100 But for some reason, people get so excited about bar charts.
00:28:08.220 Oh, yeah.
00:28:09.120 People love their bar charts, and they're misused all the time.
00:28:12.860 I think maybe it's like the easiest chart to get in your canned statistical analysis program.
00:28:17.660 But now our audience will know better.
00:28:19.900 They will know better.
00:28:21.140 All right, Kristen, let's take stock for a moment, shall we?
00:28:24.120 We have talked about the things they were not transparent about and all the red flags there.
00:28:29.840 So a little recap.
00:28:31.820 It was, they did not tell us how they defined their groups.
00:28:35.720 They dumped one of their variables, the sexy variable, and they did not report numeric summaries.
00:28:41.640 Right, yes.
00:28:42.620 All right, let's talk now about their study design and how that affected the structure of the data and the implications for that.
00:28:51.380 And this is going to bring us to correlated observations.
00:28:54.900 Oh, goody.
00:28:55.480 Yeah.
00:28:55.840 You foreshadowed that we were going to talk about this today.
00:28:58.940 I might have to get out my soapbox.
00:29:00.920 I'm surprised you don't have it all ready to go.
00:29:04.660 How do you teach about correlated observations?
00:29:07.040 Regina, I start with talking about the unit of observation.
00:29:10.100 Like, what is the thing that you are measuring?
00:29:12.520 You have to define that first.
00:29:14.080 Are you measuring a country?
00:29:15.580 Are you measuring a person?
00:29:17.040 Or maybe you're just measuring an armpit.
00:29:19.760 Right.
00:29:20.080 And each unit of observation is getting a row in the data set.
00:29:23.660 So if you're measuring a person's belly button, there's only going to be one row of data per person.
00:29:30.920 Because we only have one belly button.
00:29:32.860 What are we measuring about the belly button? 0.98
00:29:34.520 Is it we are smelling the belly button? 0.97
00:29:35.480 Not how they smell.
00:29:36.120 How much lint?
00:29:36.920 The lint.
00:29:37.280 Oh, okay.
00:29:37.720 Yes.
00:29:38.140 All right.
00:29:39.060 But if you're measuring a person's armpit separately, left armpit, right armpit,
00:29:44.900 how much lint do you have in your armpit,
00:29:46.980 there are going to be two rows of data for that person.
00:29:49.840 Right.
00:29:50.080 And if there are multiple rows of data that belong to the same person,
00:29:54.320 these are going to be correlated.
00:29:55.920 Because obviously, like if we're measuring, let's say, smell in the armpits,
00:29:59.460 The smell of your left armpit is likely to be very similar to the smell of your right armpit.
00:30:05.380 Those are not independent.
00:30:06.760 Especially if they're full of lint.
00:30:10.640 Anyway, so, okay, we are talking here about correlated observations, though.
00:30:15.000 And I think it's important that we maybe take a moment to talk about how that's different than correlated variables.
00:30:21.000 Because they kind of sound the same.
00:30:22.380 Oh, yes.
00:30:23.060 This confuses everybody because the names sound too similar.
00:30:27.140 Correlated observations arise when rows of data are related,
00:30:31.840 whereas correlated variables mean that columns of data are related.
00:30:36.660 For example, if I've measured height and weight,
00:30:39.140 these values are going to be stored in different columns,
00:30:41.540 and those columns are obviously going to be related.
00:30:44.360 But this is not correlated observations.
00:30:47.040 This is correlated variables.
00:30:49.700 Very nice. I like that one, actually.
00:30:51.620 All right, so correlated observations here.
00:30:54.540 Let's talk about why it's so important to handle them correctly.
00:30:59.460 Many statistical models assume that your observations are independent.
00:31:03.060 So if you feed correlated observations into these models,
00:31:06.180 you are violating an assumption of the model and you will get the wrong answer.
00:31:09.720 Like if our study involved 50 men but 100 armpits,
00:31:13.800 and we pretend as if we have 100 men, not just 100 armpits,
00:31:17.420 we are artificially inflating our sample size,
00:31:19.940 and the results might look more impressive than they actually are.
00:31:22.540 And a lot of people get this wrong, hence my soapbox.
00:31:26.120 Regina, also I should mention correlated observations, also sometimes called dependent observations or dependencies in the data.
00:31:34.020 Let's take this back to the sweaty t-shirt study.
00:31:36.060 We got the preliminaries out of the way.
00:31:38.520 We understand why it's important.
00:31:40.040 In this study, there were actually three sources of dependencies or correlated observations.
00:31:45.380 Three. That's complicated.
00:31:46.920 Yeah, complicated little creature there.
00:31:48.900 And let's just walk through all three.
00:31:51.440 First of all, first one, each woman rated three genetically exotic men and also three genetically similar men.
00:31:59.500 So we have six observations for each woman.
00:32:01.940 Right. We have replicates. So six rows of data per woman. How did the authors handle that? 0.94
00:32:08.320 They collapsed those six numbers basically down to two numbers.
00:32:13.340 So they took the three ratings for the three genetically exotic men and averaged those together.
00:32:20.220 and then took her ratings for the three genetically similar men and averaged those together.
00:32:26.620 And so instead of six rows per woman, now we only have two rows per woman. 0.84
00:32:30.240 Got it.
00:32:30.700 So that handles the correlated nature of the replicates, collapses three rows into one.
00:32:35.400 That's the first source of correlation.
00:32:36.940 Yeah.
00:32:37.260 Right.
00:32:37.600 Three man replicates.
00:32:39.020 Right. 0.88
00:32:39.300 Now, the second source of correlation is that we are still stuck with these two rows per woman. 0.93
00:32:46.000 Right. 0.95
00:32:46.280 Her average ODA rating for each of the two groups in there.
00:32:50.260 Right. 1.00
00:32:50.520 So this is a design where women are serving as their own controls because every woman rates both genetically similar and genetically exotic men. 1.00
00:32:57.680 So did they handle that correctly? 0.98
00:32:59.040 They did.
00:32:59.840 They used a paired statistical test. 0.98
00:33:01.880 So they compared each woman to herself. 0.98
00:33:03.940 Oh, great. 0.82
00:33:04.540 Okay.
00:33:04.800 So they handled those two sources of correlation correctly.
00:33:08.040 What's the third source, though, Regina?
00:33:09.920 Yeah, this one's trickier.
00:33:11.340 It's not as much fun.
00:33:12.420 And this is because of how they designed the experiment. 0.88
00:33:17.880 The rows of the data are also connected by the man, not just by the woman.
00:33:24.140 Right, because the men were each smelled by multiple women.
00:33:27.960 Yes. 0.66
00:33:28.560 And it was not a balanced design.
00:33:32.120 What do you mean by that?
00:33:33.200 Remember, there were 49 women and only 44 men.
00:33:38.620 So they had to use some men more than others.
00:33:41.800 Oh, right.
00:33:42.280 Some men got sniffed more than others.
00:33:45.100 The man shortage, which kind of sounds like some party.
00:33:50.220 OK, so this is super complicated.
00:33:52.220 It's making my brain hurt.
00:33:53.380 With this weird design they have, you really would have to use a fancy statistical model in order to account for that correlation by man.
00:34:02.020 So did they use one?
00:34:03.240 They definitely did not use a sophisticated statistical model.
00:34:06.840 So this is problematic, and, you know, Regina, to illustrate this, I'm going to make up an extreme case.
00:34:12.740 Sometimes you have to go to the extremes.
00:34:14.320 It makes it easier to picture in your head.
00:34:16.340 So let's imagine that in the sweaty T-shirt study, there is one man who is just genetically very exotic.
00:34:23.600 Ooh la la.
00:34:24.580 He's from Switzerland, but somehow he has a very unique lineage, very unique genetic profile,
00:34:30.520 and he pops up as the most genetically exotic man for all 49 women.
00:34:35.500 So every single woman ends up smelling his T-shirt.
00:34:39.400 So his rating is going to show up 49 times in that genetically exotic group. 0.92
00:34:44.220 Exactly.
00:34:45.200 And let's imagine also that he just smells really great.
00:34:49.660 This is a good smelling man.
00:34:51.900 Truffles.
00:34:52.800 In my mind, he smells like truffles and like ocean.
00:34:56.620 Ooh, chocolate, coffee.
00:34:59.780 Leather, tobacco.
00:35:01.120 Roses.
00:35:01.740 I love it. 1.00
00:35:03.160 Let's create a perfume and make a fortune here. 0.99
00:35:05.500 Do we have to murder him? 0.92
00:35:08.520 Like in the book? 0.96
00:35:10.980 Okay, I take it back.
00:35:12.380 Second thought.
00:35:13.580 We're just going to fantasize.
00:35:15.820 All right.
00:35:16.480 Bottom line, he smells great. 0.73
00:35:18.120 It has nothing to do with his genes or the match of the genes with these different women.
00:35:22.800 He is just an objectively good-smelling guy.
00:35:25.440 And most women who smell him are going to rate him highly.
00:35:28.500 That 8, 9, or 10 out of 10, basically.
00:35:31.140 This means that every woman in the study has at least one high score in her genetically exotic pile.
00:35:36.680 He's averaged into the exotic scores 49 times, more than anyone else.
00:35:41.240 Outsized influence this one guy has.
00:35:43.060 Yes, this one man could be driving the entire finding.
00:35:46.380 With him in the data set, maybe you find a big difference between genetically exotic and similar men.
00:35:51.180 Yeah, but if you were to drop him, maybe it just goes away entirely.
00:35:54.660 Yeah, he could be driving the entire effect.
00:35:56.800 And obviously the issue is we are counting one man as if he represents 49 independent men.
00:36:03.320 When we ignore the correlated nature of the observation, we end up over counting him and this can skew the results.
00:36:09.340 This is why they should have used a more sophisticated model here or even better, a better study design.
00:36:14.840 Of course, I should clarify this is an extreme example.
00:36:19.440 Probably not what happened in the study.
00:36:21.100 Probably not what happened in the study.
00:36:22.400 Right, more subtle.
00:36:23.020 But it does show why it's important in there.
00:36:26.500 And the point is they had a weird data structure that they did not really deal with.
00:36:31.060 This example is meant just to drive the intuition.
00:36:33.560 We actually don't know because they weren't transparent.
00:36:35.680 We don't know how many men were rated by how many women.
00:36:38.440 We don't know what effect this had on the results.
00:36:40.340 But we do know that they failed to account for something that could be important.
00:36:44.020 Right, right.
00:36:44.820 So, Regina, tell me more about the results.
00:36:46.580 Right, the results.
00:36:47.580 We're ready for the results.
00:36:48.900 First of all, remember they never reported on sexiness.
00:36:52.980 Right.
00:36:53.420 They presented results just for how intense the body odors were and how pleasant they were.
00:37:00.360 And for intensity, there were actually no significant effects at all.
00:37:05.220 I know.
00:37:05.640 So I'm just going to talk about pleasantness.
00:37:07.380 Okay.
00:37:07.760 Simplifies things.
00:37:08.680 They presented their results in a weird way, but based on what I could see of the results,
00:37:14.140 it looks like overall there were actually no significant differences
00:37:19.280 in how women rated the pleasantness of exotic or similar men.
00:37:23.840 Wait, wow.
00:37:24.960 So no effect.
00:37:26.880 But why then does everyone cite this paper, I guess including me,
00:37:32.000 if they didn't actually find a difference?
00:37:33.800 I know.
00:37:34.240 Good question.
00:37:35.400 And the answer to that is to get the results that they highlighted,
00:37:39.820 you know, the exciting results in the discussion,
00:37:41.640 they did two weird things with their data.
00:37:44.220 Uh-oh.
00:37:44.760 I'm foreseeing some data shenanigans.
00:37:48.280 I was right.
00:37:48.860 First one, they analyzed their data from two different perspectives, meaning they looked at how women rated the smell of men in the two genetic groups. 0.57
00:37:59.740 Right. 0.90
00:38:00.060 That's the women's perspective. 0.96
00:38:02.300 That makes sense. 0.94
00:38:03.080 That's kind of what we talked about already.
00:38:04.360 Talked about.
00:38:04.780 But they also analyzed the data from the man's perspective, meaning they recalculated the averages for each man.
00:38:13.720 Okay.
00:38:14.080 Explain.
00:38:14.540 What do you mean by that?
00:38:15.420 I know.
00:38:15.900 I know.
00:38:16.160 But let's do a hypothetical like you did before.
00:38:19.040 Let's say one of the men was rated by five genetically exotic women, different MHC genes, and three genetically similar women.
00:38:28.020 Sure.
00:38:28.620 He would have now eight rows of data that belonged to him.
00:38:31.780 So we're going to take those five exotic scores and then the three similar scores and compare those two averages just like we did.
00:38:39.780 So we kind of inverted the whole thing.
00:38:41.840 Right.
00:38:42.220 Like a man is serving as his own control now.
00:38:45.040 Oh, that is weird.
00:38:46.160 It is weird, yeah. 1.00
00:38:47.160 Because, I mean, the study was set up about the women. 0.95
00:38:49.540 It's the women smelling the men. 0.79
00:38:50.780 This doesn't make a lot of sense. 0.87
00:38:52.100 And it's not going to give us the same answer as we do it from the women's perspective. 0.99
00:38:57.120 Exactly. 0.55
00:38:58.160 And, okay, ultimately, the results were more exciting, only in one direction, only from the man's perspective.
00:39:05.000 Oh, I see.
00:39:05.440 So that's what they highlighted.
00:39:07.120 Of course.
00:39:07.940 I know.
00:39:08.620 I know.
00:39:09.000 So that leads us actually into the second weird.
00:39:11.160 Oh, there's another one.
00:39:12.060 There's another one.
00:39:13.340 They did not stop.
00:39:15.720 They did something I found quite suspicious. 1.00
00:39:19.240 They split the women into two groups.
00:39:21.380 Women on hormonal birth control, the pill, and women not on the pill.
00:39:26.240 And then they analyzed each group separately, reported each group separately.
00:39:31.240 And when you split the women into two groups like this and only when you analyze it from the man's perspective, that's when you get the exciting results. 0.69
00:39:40.380 Okay, but the women are on the pill, not the men. 0.59
00:39:42.840 So how do you do this from the point of view of the men?
00:39:45.020 The men are not on the pill.
00:39:46.140 I know.
00:39:46.720 I told you it was a mess.
00:39:48.160 Okay.
00:39:48.260 This is kind of a mess.
00:39:48.900 Well, first of all, what were the exciting results?
00:39:51.100 All right.
00:39:52.380 Women not on the pill preferred the smell of men who had different MHC genes.
00:39:57.600 Okay.
00:39:57.780 That was the expected result.
00:39:59.340 We were expecting that.
00:40:00.020 Uh-huh.
00:40:00.400 It went in the right way. 1.00
00:40:01.360 But women on the pill had the reversed preference. 0.99
00:40:05.520 Oh.
00:40:06.060 They preferred men with similar MHC genes.
00:40:10.320 So that preference completely flipped.
00:40:11.940 The opposite of what was expected.
00:40:14.520 All right, Regina, I'm wondering, though, was this stratification, dividing it up by women on the pill versus not on the pill, did they plan to do this ahead of time?
00:40:23.140 Yeah, right.
00:40:23.900 They were not transparent about it.
00:40:25.640 Shocking.
00:40:26.040 Not surprising.
00:40:27.280 I suspect it was not.
00:40:28.640 Okay.
00:40:28.920 They didn't mention anything about it in the intersection or citing literature about this.
00:40:33.860 And also, part of that, though, the clue is the sample sizes.
00:40:37.920 Oh, what was the sample size? 0.87
00:40:39.320 Out of the 49 women, how many were on the pill?
00:40:41.380 Yeah, 18 on the pill, 31 not on the pill.
00:40:45.040 Right, so I see what you're getting at, because if they had planned this, this isn't an experiment,
00:40:48.860 and if they cared about pill versus not pill, they would have gone out and recruited about the same number in each of those groups
00:40:54.060 to maximize the statistical efficiency.
00:40:56.400 So this makes me suspicious as well, Regina.
00:40:59.400 This is like you really have to twist the data through a lot of gymnastics to get some exciting result,
00:41:05.260 which makes me think that it might not be robust.
00:41:08.060 And this is one of the things that you're supposed to do as a statistician.
00:41:10.660 You're supposed to try to break your analysis.
00:41:13.020 You are.
00:41:13.440 Throw it on the floor and see if it shatters.
00:41:15.680 Because if your result only holds up, when you twist the data in this very particular configuration, right, it's probably not robust.
00:41:25.040 It's probably just a statistical artifact.
00:41:27.180 Might be an artifact.
00:41:28.220 I am wondering if there were some post-hoc analyses going on.
00:41:32.060 And by post-hoc, after the fact, meaning they were not pre-planned, these analyses.
00:41:36.640 Maybe when they analyzed everything as just one big group, nothing came out significant.
00:41:41.380 Right.
00:41:41.900 I mean, this is what I'm thinking, too.
00:41:43.700 This is what might have happened.
00:41:45.300 We didn't find anything.
00:41:46.440 So let's start twisting our data and looking for some.
00:41:50.780 Can we pull something out of the data?
00:41:52.780 Split it by pill and not on the pill.
00:41:55.800 Right.
00:41:55.960 It often ends up to be just a statistical artifact.
00:41:59.120 Yeah, something that is just a fluke in there.
00:42:01.740 And, of course, we're going to drop sexiness.
00:42:03.640 Oh, right.
00:42:04.240 Yeah, they drop.
00:42:04.840 So I'm really curious about the sexiness variable, that rating, because why did they drop it?
00:42:10.160 They gave us the intensity one, which was not significant.
00:42:13.460 Right, I know.
00:42:13.900 Was it like the sexiness went in the opposite direction or something so they couldn't make a good story?
00:42:18.820 That is, yeah.
00:42:19.540 We will never know.
00:42:20.840 That was 1995.
00:42:22.900 But we know how these things go.
00:42:24.860 Right.
00:42:25.160 Yeah, we don't actually know what happened behind the scenes.
00:42:27.200 We don't know what the authors did.
00:42:28.780 But the two of us have enough experience that we can kind of guess that something like this probably went on behind the scenes.
00:42:33.740 Yeah, we know what happens in the dark recesses of the research basement.
00:42:39.460 Shenanigans.
00:42:40.440 It's not pretty.
00:42:41.340 So I think that leads us very nicely, actually, to talking about their conclusions in their discussion section.
00:42:48.420 And you are going to get a kick out of this.
00:42:50.620 I can't wait.
00:42:54.940 You want to impress them on a first date, but also play it cool.
00:42:59.280 So what do you do?
00:43:01.660 I'm Rufy Thorpe, and I wrote and read a real love story about a hinged couple that navigated exactly that.
00:43:07.740 Listen to the free audiobook now.
00:43:13.380 We are talking about the sweaty t-shirt study, and we're about to hear the conclusions of that study.
00:43:18.880 I think you're going to like this, Tristan.
00:43:20.800 Really?
00:43:21.460 And by like, I mean hate.
00:43:23.860 So I might have some criticisms is what you're saying.
00:43:26.040 Maybe a couple.
00:43:26.780 All right, give me.
00:43:27.320 statement. Let me read you the first line of the discussion. The contraceptive pill seems to have
00:43:35.020 a strong influence on odor preference. Oh, wow. So that is an overstatement. They actually don't
00:43:42.660 have data to support that statement. And this is a little subtle, but they haven't compared the
00:43:48.460 women on the pill directly to the women not on the pill, which is the analysis you would need to do
00:43:53.240 to make that statement.
00:43:54.360 And it's a little bit of a subtle statistical point.
00:43:56.760 It's what's called a test of interaction.
00:43:58.160 We'll discuss it in a future episode maybe,
00:43:59.920 but suffice it to say,
00:44:01.660 they don't have data to support that.
00:44:02.940 It's the first sentence of their discussion.
00:44:04.540 The first sentence.
00:44:05.700 Okay, next line.
00:44:07.100 This indicates that steroids,
00:44:09.480 which are naturally released during pregnancy, 0.96
00:44:11.460 could change body odor preferences,
00:44:13.360 leading to a preference for odors
00:44:14.780 which are similar to those of relatives.
00:44:16.620 Okay, wait a minute. 0.95
00:44:17.960 How did we get to pregnant women?
00:44:19.520 I thought the women in the study were not pregnant, right? 0.99
00:44:22.240 They were not pregnant.
00:44:23.240 They were on the pill, but being on the pill is not the same as being pregnant. 0.87
00:44:26.820 And I can tell you that from personal experience, not the same. 0.95
00:44:30.360 Yeah, they're kind of, they're going off there.
00:44:33.060 Then they go off on pregnant mice, actually.
00:44:35.400 They have a few lines about pregnant mice and what the pregnant mice like to do.
00:44:39.560 So we have wandered from women on the pill all the way over to pregnant mice.
00:44:44.500 This is what I get really upset about often when I'm reading discussion sections.
00:44:48.480 Authors wander too far from their data.
00:44:50.480 They are like wandering out into the wilderness here at this point.
00:44:53.240 Yeah, I think these people might be galloping into the wilderness.
00:44:56.820 Running into the wilderness, yes.
00:44:58.220 Way far from their data, straying from their data, yes.
00:45:01.580 Okay, then they go on to say,
00:45:04.160 therefore, the contraceptive pill seems to interfere with natural mate choice.
00:45:10.180 Oh, no way.
00:45:12.780 Wow.
00:45:13.380 That's a good one, right?
00:45:14.180 There's a little anti-pill here.
00:45:15.700 Is this like the political agenda behind this? 1.00
00:45:18.520 Pills are destroying the very fabric of marriage. 1.00
00:45:21.480 Well, and also— 0.84
00:45:22.640 women today mate choice was not an outcome here it was how pleasant the men smelled so we've gone
00:45:29.100 from the pleasantness of the men right not even sexiness right not even sexiness but to mate
00:45:34.000 choice right yes because we're all running around saying oh that man is pleasant i'm going to marry
00:45:37.880 him he has a pleasant smell so that's enough right but yet this study gets cited all the time
00:45:44.380 right you said it's widely cited and i apparently might have cited it myself when i said that
00:45:48.920 Studies show, please tell me there's other data that support my line, studies show, or did I just get it wrong?
00:45:55.040 First of all, I did the same thing.
00:45:56.280 I referenced this paper when I was writing about the pheromone dating parties for the LA Times.
00:46:01.980 So you and I are both doing a little, you know, mea culpa here.
00:46:06.560 There have been follow-up studies, but before that, I'd like to dwell on the splash that this made.
00:46:12.240 Did this get some media attention?
00:46:13.960 It did.
00:46:14.820 It did. In fact, one in New Scientist by our colleague, Peter Aldis.
00:46:19.760 Oh, well, Peter was my professor at UC Santa Cruz in the science writing program,
00:46:24.320 although I completely cut his class.
00:46:26.620 You cut his class?
00:46:27.660 Yeah, I was on a boat in Antarctica for that quarter, so I skipped his class entirely.
00:46:32.220 That is such a great story. We're going to have to save that one.
00:46:34.840 Another day, we'll talk about that.
00:46:36.380 But question for you, was Peter credulous or skeptical of this study when he wrote about it?
00:46:41.920 Credulous, I'm afraid. Let me read you his lead.
00:46:44.760 Women are attracted to the odors of men who have different immune system genes from their own, 0.93
00:46:48.940 but only if they're not taking the pill. 0.90
00:46:50.780 Okay, so he bought it hook, line, and sinker.
00:46:52.700 So I didn't miss anything by cutting his glass.
00:46:56.320 Next time I see him, I'm going to tell him that.
00:46:58.620 I'm kidding, I'm kidding.
00:47:00.460 But he did add this bit of information, investigative journalism.
00:47:04.820 I think you'll like it.
00:47:05.640 He was talking about the reversal of preferences on the pill. 0.71
00:47:08.880 Oh, where the women on the pill like the similar men better than the dissimilar men, yes.
00:47:13.300 Here's his quote.
00:47:14.360 Wiedekin is still mystified by this result.
00:47:17.620 Well, I am mystified.
00:47:18.540 I am mystified by that quote because in his discussion section,
00:47:22.460 he sounds pretty sure of himself.
00:47:23.720 He did not sound mystified.
00:47:25.000 Yeah, interesting.
00:47:25.640 Yeah, so it just adds to my suspicion they didn't plan this.
00:47:29.300 Because if they had planned this pill,
00:47:31.400 then why would they be mystified at the result?
00:47:33.600 If he was mystified by it, he clearly didn't expect it ahead of time,
00:47:36.880 so it wasn't a pre-planned analysis,
00:47:38.680 adding to our suspicion that it's post hoc.
00:47:41.000 At least Peter didn't gush, though.
00:47:42.640 Here's one story in the news section of an ecology journal.
00:47:46.140 And they were downright giddy.
00:47:47.500 Oh, really?
00:47:48.120 Let me read you a quote. 0.91
00:47:49.620 A mate chosen by a woman on the pill might smell sweet at the time, 0.97
00:47:53.260 but when she goes off the pill to reproduce, 0.66
00:47:55.780 she might likewise go off his scent.
00:47:58.320 Could the rise in divorce rates be correlated with mate choices made on the pill?
00:48:03.020 The rise in divorce rates?
00:48:04.480 Wow. 0.80
00:48:05.180 So we've gone from a few women on the pill in this weird analysis 0.99
00:48:08.900 to now we can explain why divorce is on the rise.
00:48:11.840 Wow.
00:48:12.640 That is wandering very far from the data.
00:48:16.240 Yeah, they're like ocean maybe.
00:48:17.660 Yeah, out in the ocean in the middle of Antarctica.
00:48:21.880 Luckily not, everyone loved it.
00:48:24.280 Oh, I'm glad to hear that.
00:48:25.260 Yep, yep.
00:48:25.860 A year later, there were two researchers who wrote a letter to the editor in the Ecology Journal.
00:48:30.980 Okay, fine.
00:48:32.580 Phil Hedrick and Volker Luska.
00:48:35.900 Okay.
00:48:36.280 And they picked up on some of the same things we did.
00:48:38.720 Oh, yay.
00:48:38.980 It was really good.
00:48:39.900 They said, experimental design, data need more scrutiny.
00:48:43.500 Sample size is too small.
00:48:45.080 They also picked up, like we did, on how the results differed
00:48:48.540 depending on whether you're using it from the point of view of men or women. 0.90
00:48:51.060 Yeah, that weird thing, yep.
00:48:52.140 And they said, you need to explain that.
00:48:53.860 And they pointed out, you know, humans are not mice.
00:48:57.980 Right, yes, but true, true.
00:48:59.800 With mice, it's easy because you can have a cone and you can manipulate genes.
00:49:03.660 So making inferences based on the mice studies doesn't make a lot of sense.
00:49:07.080 Well, that's great.
00:49:08.320 I know.
00:49:08.580 Did people read their letter to the editor?
00:49:11.320 And, well, clearly not, since this is a widely cited study.
00:49:14.880 I'm guessing it kind of got buried, as sometimes happens with criticisms, yep.
00:49:18.440 You know who did read it in response, however, was Wiedekind and one of the other authors of the T-shirt study.
00:49:24.260 Right, I imagine they weren't happy about it.
00:49:25.760 Yeah, they were not happy.
00:49:27.360 It's kind of funny to read.
00:49:29.000 Reading between the lines, not a quote, but they kind of accused those letter writers of fundamentally misunderstanding statistics and science.
00:49:35.960 Well, they criticized us, so of course they misunderstood everything.
00:49:39.760 And we defended their study design, which doesn't really have a lot in its defense.
00:49:46.200 Right, exactly, yes.
00:49:47.940 And they seemed a little miffed at how the letter writers were questioning their data analysis.
00:49:52.360 So they said, okay, we're going to send you our data so you can analyze it.
00:49:55.900 They sent the data.
00:49:56.920 I know.
00:49:57.340 Did the letter writers ever do anything with that data?
00:50:00.940 No.
00:50:01.180 Can we get a hold of that data?
00:50:02.460 I know.
00:50:03.020 I couldn't find that they published.
00:50:05.300 on it. I searched and searched, so I emailed them. Oh, you did? I emailed them. They're both
00:50:10.240 emeritus faculty now. Right, this was a while ago. Oh, this was 1996 when you and I met. Did they
00:50:15.120 remember this? Oh, they did. One's in Arizona, another is in Denmark. And I said, hey, you know,
00:50:21.660 it's 30 years later and all, but great letter to the editor back in 96. You know, I think if I'm
00:50:27.520 an emeritus professor someday and somebody says, that article you wrote 30 years ago,
00:50:31.860 I loved it. I'm happy.
00:50:34.240 This is happy. And it was genuine
00:50:36.180 because it was really well done.
00:50:37.980 And I said, and then, by the way, do you still have that data
00:50:39.900 that we sent you? Did they have the data?
00:50:42.300 They did not have the data. One said,
00:50:44.260 oh, I don't remember if we analyzed it
00:50:46.160 or couldn't figure out how to analyze it.
00:50:47.900 The guy in Denmark is like,
00:50:49.920 oh, I'll go look for it.
00:50:51.600 I mean, it's 30 years later.
00:50:53.740 They didn't have cloud storage.
00:50:55.860 Back then. Yeah. So we ended up
00:50:57.880 having this really nice email conversation
00:50:59.540 on a Saturday night, you know,
00:51:01.080 Okay, about the study.
00:51:03.060 Talking about, it was interesting, the geneticists,
00:51:05.040 so talking about where this whole field has gone since then
00:51:07.640 and what Wiedekin has been up to.
00:51:09.580 Ooh, what has he been up to?
00:51:10.680 So he published two more studies on the topic,
00:51:13.800 but mostly he works on fish biology now, apparently.
00:51:16.800 Oh, well, maybe that's a good thing.
00:51:20.260 So they sent me the two papers.
00:51:23.480 The 1997 paper was kind of a replication.
00:51:28.160 Oh, well, good for him for trying to replicate the findings.
00:51:30.620 Yes.
00:51:31.060 Okay, good.
00:51:32.100 Okay, study design, a little different.
00:51:34.000 Okay.
00:51:34.360 He worked with six t-shirt wearers.
00:51:37.200 Only six this time.
00:51:38.300 Only six, a mixture of males and females.
00:51:40.360 Oh, are we smelling both ways then?
00:51:42.260 Men smelling women, women smelling men?
00:51:43.840 121 sniffers, smellers, also a mixture of male and female.
00:51:48.940 Oh, interesting.
00:51:49.660 Okay.
00:51:50.360 And some of the same t-shirt wearers and sniffers as in his 95 paper, his original paper.
00:51:57.100 Oh, some of the same study subjects.
00:51:58.500 Some of the same people.
00:51:59.400 Oh, that is a red flag for me.
00:52:02.280 I always worry about when I see that because that makes me think that they're on the friends, family, staff, and grad students plan meeting.
00:52:10.600 Hey, I'm just going to pick study subjects from whoever I can find sitting, you know, at the desk next to me.
00:52:16.760 And that's really problematic because that is not the way that we should select study subjects.
00:52:21.640 The people who are working in the lab, for example, are the grad students.
00:52:25.640 They may have a vested interest in how those results come out.
00:52:28.300 And also, it's just not a really good generalizable sample.
00:52:31.100 It's not a representative sample.
00:52:32.820 So, don't like that right away.
00:52:34.400 So, not only that, but it had a lot of statistical issues as well.
00:52:38.620 Same, similar to the other paper.
00:52:40.100 Yeah.
00:52:40.680 So, just want to give one example.
00:52:42.880 They found no significant effect for T-shirts being rated by women on the pill.
00:52:48.040 Okay.
00:52:48.400 No significant effect for T-shirts being rated by men.
00:52:52.220 No significant effects for T-shirts being rated by women not on the pill.
00:52:56.440 So, none of it replicated then?
00:52:58.120 No, but why stop there, Kristen?
00:53:00.780 Just because you have three non-significant results,
00:53:03.560 let's start combining subgroups.
00:53:05.260 Oh, no.
00:53:05.860 Yeah, right.
00:53:06.640 Because if you combine the men and the women not on the pill,
00:53:11.180 then you get significance.
00:53:12.500 But wait a minute. 0.93
00:53:13.420 Why in the world would you combine men with women not on the pill? 0.97
00:53:18.040 Like women who are menstruating and ovulating? 0.96
00:53:20.140 How is that combinable with men? 0.99
00:53:22.420 Right. 1.00
00:53:22.700 If anything, they'd be combining the women
00:53:24.900 or maybe the women not on the pill.
00:53:26.420 Yes, that's the only way they could get significance, I guess.
00:53:30.040 I know.
00:53:30.680 They dropped sexiness.
00:53:32.180 Oh, again, no sexiness.
00:53:33.460 Bar graphs.
00:53:33.940 Oh, all the same problems.
00:53:34.660 Some of their analyses had even smaller numbers, smaller sample sizes.
00:53:38.880 Just a little bit of a dumpster fire.
00:53:40.520 It sounds like it.
00:53:41.520 Yes.
00:53:42.000 Okay, so it really, that's their quote-unquote replication.
00:53:45.420 But if you're hoping to make the whole thing more robust by pointing to that replication, you've gone awry.
00:53:51.940 Okay, but I'm sure that other people have tried to replicate this result since 1995.
00:53:56.080 What are some of the other studies that have been done on this topic since then?
00:53:59.020 A number of people have tried to do this over the 25 years, so many of them,
00:54:04.240 that now people are able to do a meta-analysis where they combine all the results.
00:54:09.100 So do you want to explain a meta-analysis?
00:54:10.640 Sure. So meta-analysis just means you take a bunch of papers,
00:54:13.320 and you're just taking the summary data from those papers,
00:54:15.420 not necessarily the underlying original data,
00:54:17.420 and combining it to try to get some sense of the overall effect.
00:54:20.920 So maybe one study found something, another didn't,
00:54:23.420 but science isn't just about one paper.
00:54:25.020 So let's pool all that data and see, overall, do we find that women prefer the genetically dissimilar men?
00:54:31.640 So there was a really nice one in 2020, a big meta-analysis, odor preference and genetics, and they found 10 nice studies to include.
00:54:40.040 And what did they find?
00:54:40.800 First, I want to give a nice call out to them because this Wiedekin 1997 replication paper—
00:54:46.960 It was included in this meta-analysis.
00:54:48.660 It was included, but they didn't fall for this combining the subgroups thing. 1.00
00:54:53.780 When they extracted the data, they kept the women on the pill
00:54:57.160 or not on the pill separate from the men.
00:54:59.000 And they had to use a little digital tool to be able to get the results out.
00:55:02.420 Oh, they extracted the data from those bar charts too.
00:55:05.480 So it was well done.
00:55:06.420 I just wanted to call them.
00:55:06.720 So we have 10 studies in this meta-analysis.
00:55:08.480 Two of those are the Wettekin studies, the 1995 and the 1997,
00:55:11.680 but they are calling the 1997 results not significant.
00:55:15.220 Okay, they pooled the data.
00:55:16.860 What did they find?
00:55:17.340 Yeah, nothing, nada, no evidence.
00:55:20.000 So no finding, no evidence for this claim.
00:55:22.340 The only significant effect out of those 10 studies was Wiedekin 1995, that original study.
00:55:28.540 That was it.
00:55:29.560 So that original study found a significant effect.
00:55:31.780 The 1997 study, they're calling it out and saying, actually, that was a failure to replicate as we looked at it.
00:55:38.000 And then eight other studies found nothing.
00:55:40.040 So there's nothing here.
00:55:41.960 There's nothing here.
00:55:43.320 I'm really sad about this.
00:55:44.420 I'm kind of crushed.
00:55:44.960 You wanted this to be true.
00:55:46.200 I did.
00:55:46.940 This is such a fun result.
00:55:48.640 People love to talk about it.
00:55:50.380 But that first study was just so bad that it's not surprising that it didn't replicate, actually.
00:55:56.760 They got lucky.
00:55:57.660 Yeah.
00:55:57.980 And they published that, and people could not replicate.
00:56:01.380 Now, we're not saying that smell doesn't influence mate choice, right?
00:56:05.100 Our claim that we're looking at today was that it's mediated through these HLA genes.
00:56:09.420 But, of course, people still do choose mates based on smell.
00:56:12.460 Mm-hmm.
00:56:13.140 Mm-hmm.
00:56:13.660 In fact, some studies suggest—
00:56:15.820 Uh-oh.
00:56:16.260 So, we don't know if we'd write or not.
00:56:17.760 But studies suggest that people are doing this, that our body odor is reflecting things like our personality, our mood at the time.
00:56:26.080 Yes, our health, our general health.
00:56:27.300 It certainly reflects, like, did you eat garlic last night?
00:56:30.340 But even beyond that, so maybe we can do an episode on that.
00:56:34.040 We should, yeah.
00:56:35.140 We might want to dig into some of those studies that suggest, see if they're real, but I'd be curious to do that.
00:56:40.720 All right, Regina, this is the part of the episode where we're going to wrap everything up,
00:56:44.360 and we're going to rate the strength of the evidence for the claim that we're looking at
00:56:47.560 today. And the claim is that women prefer the smell of men who are genetically dissimilar to
00:56:54.740 them. And how do we rate strength of evidence on this podcast? We use our trademarked, highly
00:57:00.300 scientific smooch rating scale, one to five smooches, kind of like Amazon stars. One smooch
00:57:06.620 means little or no evidence for the claim. And five smooches means very strong evidence for the
00:57:11.820 claim. What say you, Regina? Kiss it or diss it? I'm going to need to diss this one. I'm going to
00:57:18.760 give it one smooch. I am so disappointed in this original study. It was way weaker than I had
00:57:27.500 expected or hoped, and there's just nothing there. One smooch. Yeah, I have to agree with you,
00:57:32.660 Regina. This is a clear one smooch from me, too. This is so widely cited that you would think that
00:57:38.060 there's something behind it. So it's really disappointing that this study is so bad. And
00:57:43.420 it hasn't been replicated at all despite attempts to do so. Yeah. So I'm going one smooch, no
00:57:48.260 evidence. Maybe this could be true in mice, but not humans. I might still use it as a screening
00:57:53.820 tool for my men on dates, but nothing to do with the genetics. Yeah. You might still throw a
00:57:59.160 pheromone party, but not to do with the genetics necessarily. The other thing we like to do in
00:58:04.440 podcast is because we're not just talking about this claim that we're looking at today. We're
00:58:09.460 talking about how to evaluate evidence for a claim like this. And so we'd like to give some
00:58:14.280 methodologic morals. And these are a little like Aesop's Feeble morals. So Regina, do you have a
00:58:20.420 methodologic moral for us today? I have many. Oh, I'm going to let you have more than one today.
00:58:26.000 So I'm going to pick one. It's a small point in today's story, but it's something I really want
00:58:31.120 people to remember. So repeat after me. Bar charts are not for numerical data. Bar charts are not for
00:58:38.720 numerical data. Got it. Good job. Good job. Can I have a sub moral? Of course. Yes. Always report
00:58:45.860 your summary statistics. Don't make readers guess by looking at your crappy charts. Oh, hallelujah 0.97
00:58:50.480 to that one. Oh, very nice. Can I have one more? You get one more today. Absolutely. Okay. Don't
00:58:57.020 ago ranting about off-topic crap in your discussion. Oh, yeah. Agree with that one too, 0.98
00:59:02.440 Regina. Okay, you. So mine is going to be those who ignore dependencies in their data are destined
00:59:08.620 for flawed conclusions. Oh, I love it. Yeah, it's true. Something to always keep in mind when you're
00:59:14.340 analyzing data. Dependent data. Well, this has been a fascinating episode. I've learned so much.
00:59:20.040 Thanks so much. Thanks, Kristen.
00:59:27.020 you matched on hinge you're vibing then her energy completely changes what do you do
00:59:37.380 i'm raven smith and i wrote and read a real love story about this exact
00:59:41.880 sitch listen to the free audiobook now