Why Men & Women Show Attraction Differently | Dr. Michael Platt & Dr. Andrew Huberman

00:00:00.000 | - What are the data on how females signal,

00:00:05.000 | let's just say, testosterone, estrogen,

00:00:09.680 | and other relevant hormones, and for males as well,

00:00:14.680 | what are the external signals or behavioral signals?

00:00:18.860 | - Yeah, so that's a really important point that you made

00:00:21.760 | because they both, those things go together.

00:00:24.760 | So it's been most controversial for females,

00:00:28.300 | but in my view, the data is pretty clear,

00:00:30.960 | and it aligns, I think, with our own intuitions

00:00:35.020 | just from daily life, which is,

00:00:37.440 | well, some things are apparently not consciously perceptible.

00:00:42.440 | It's hard to report, but through studies

00:00:46.600 | where you just ask males for like,

00:00:48.520 | okay, how attractive is this woman, or et cetera,

00:00:51.180 | that there are changes in the face, for example,

00:00:55.040 | and that's been one argument is that,

00:00:56.960 | this is gonna sound funny, but that the signals

00:00:59.520 | that in non-human primates are in the rear are,

00:01:03.600 | because we're walking upright, you can't see that really,

00:01:06.080 | so now it's kind of in the face,

00:01:08.300 | and so these changes that happen,

00:01:10.960 | that the ovulatory cycle is reflected in the turgidity,

00:01:15.120 | how tight the skin is in the face,

00:01:16.760 | because it gets a little plumper and a little bit redder,

00:01:20.080 | and we may not be consciously aware of that,

00:01:22.520 | but that it's there, right, and it shows up

00:01:26.160 | in sort of preference data when you ask

00:01:29.200 | heterosexual males, how attractive is this woman, et cetera,

00:01:32.440 | so that seems to be the case.

00:01:35.040 | And also behavioral, so sort of flirtatious behavior--

00:01:40.040 | - Increases around the time of ovulation.

00:01:45.240 | - Yeah, yeah, yeah, yeah, there is a classic study

00:01:48.840 | that exotic dancers, strippers,

00:01:54.760 | would actually get bigger tips, more tips,

00:01:57.520 | when they were ovulating than when they're not ovulating.

00:02:00.840 | - Interesting.

00:02:01.680 | - So there may be--

00:02:02.640 | - And it could be by virtue of their behavior,

00:02:06.120 | but it could be the way they dance, proximity to the,

00:02:09.720 | what I guess the observers, clients,

00:02:11.440 | whatever you call them.

00:02:12.400 | - I don't recall that being quantified,

00:02:15.360 | but it suggests that there's a latent signal there.

00:02:21.160 | And that men are unconsciously processing this.

00:02:26.160 | They're not saying, "Oh, her cheeks are particularly

00:02:29.360 | "plump and red right now."

00:02:32.920 | But that if you measure their ratings

00:02:37.740 | or scores of attractiveness, when she's ovulating,

00:02:41.880 | it's these features that might be drawing out that response.

00:02:45.080 | - Correct, we can take this back to the monkey porn studies,

00:02:47.520 | which was our first real foray into trying to quantify

00:02:52.520 | the value of various kinds of social information

00:02:59.960 | for guiding decisions.

00:03:02.160 | And we already came into this with a sense that like,

00:03:06.080 | yeah, things like status, physical prowess, mating status,

00:03:10.760 | are you, you look like a good mate, bad mate,

00:03:13.640 | are you in mating condition, et cetera.

00:03:17.600 | And so when you think about that,

00:03:20.880 | how do you ask a monkey that question?

00:03:23.440 | You could ask them, they're not gonna tell you

00:03:25.080 | 'cause they can't talk, but you have to develop

00:03:28.000 | a behavioral way to elicit that.

00:03:30.000 | And so what we did, I think it was pretty clever,

00:03:33.520 | was to riff on the studies that I had already done

00:03:38.520 | looking at varying the expected value of two options.

00:03:41.780 | So this was the work I did as a postdoc with Paul Glimcher,

00:03:45.200 | where we revealed economic signals in the brain,

00:03:50.200 | in the parietal cortex, an area between

00:03:53.160 | where visual signals come in and where you make a choice

00:03:56.640 | to make a behavioral response.

00:03:59.240 | And we varied, like in this case,

00:04:02.560 | monkeys don't work for money, though they work for juice.

00:04:05.440 | Okay, it's been actually, it's really fun.

00:04:07.160 | You spend a lot of time figuring out what juice

00:04:08.600 | they really love best.

00:04:10.120 | And then economically, you would vary the size

00:04:13.540 | of the juice reward that each of the two offered,

00:04:16.420 | or its probability while maintaining size constant,

00:04:19.780 | that when you combine those, you multiply those together,

00:04:21.940 | you get expected value.

00:04:23.180 | That's the first model of economic decision-making

00:04:26.260 | that was really ever developed, right?

00:04:27.420 | You compute the expected value, different options,

00:04:29.140 | you choose the one that has the highest value.

00:04:31.340 | It doesn't work all the time, but it's sort of a rough proxy

00:04:34.700 | and we showed that, yeah, neurons in the parietal cortex

00:04:36.700 | signal that.

00:04:37.520 | Monkeys are good economists.

00:04:38.360 | They choose the one that has a higher expected value.

00:04:40.460 | Okay, so now take that experiment.

00:04:42.860 | I'm gonna have monkeys choosing between two options

00:04:46.100 | that vary in how much juice they pay out.

00:04:48.580 | But I'm also gonna pop up a picture

00:04:51.540 | when they choose one of them, okay?

00:04:54.300 | And they don't know what picture's coming up,

00:04:56.060 | but the picture's gonna be, it could be a nothing burger,

00:04:59.700 | just like some gray square, it doesn't mean anything.

00:05:02.420 | Or it could be the perineum of a female,

00:05:06.420 | if it were males that we were studying.

00:05:07.620 | We did this with males, sorry,

00:05:09.000 | females making choices eventually as well.

00:05:11.160 | Could be face of a dominant male,

00:05:13.160 | face of a subordinate male, face of female, et cetera.

00:05:15.400 | - What's the equivalent of the swollen taint

00:05:18.860 | of a female monkey for, if you reverse the experiment

00:05:23.360 | and it's the female monkey who's making a choice

00:05:25.240 | about male monkeys, what do they find really attractive

00:05:28.520 | in a male monkey?

00:05:29.360 | - Yeah, so it's the taint of the male monkey

00:05:31.160 | 'cause it's providing a signal about how much-

00:05:33.560 | - Monkeys looking at taints about monkeys.

00:05:34.400 | - Yeah, how much testosterone is circular,

00:05:37.980 | that they've got on board basically,

00:05:39.780 | which is a good predictor of their status.

00:05:43.100 | It's a good predictor of their fighting ability,

00:05:45.340 | all that kind of stuff.

00:05:46.180 | And if you're a female,

00:05:47.300 | that's a reasonable kind of choice to make

00:05:50.620 | 'cause if you have male offspring

00:05:51.980 | and females are predisposed to choose that,

00:05:53.860 | then your male offspring are gonna do pretty well.

00:05:56.940 | So that's what we did.

00:05:58.160 | And we varied how much juice.

00:06:00.020 | So sometimes monkeys would get paid,

00:06:01.940 | they'd have to give up juice to see the pictures.

00:06:04.340 | Sometimes they get paid more to see the pictures.

00:06:06.980 | And what we did then is we construct a choice curve

00:06:10.400 | and we use the differential.

00:06:12.260 | If it's not 50/50, if it slides one way or the other,

00:06:16.020 | it tells us that monkeys are paying X amount

00:06:20.180 | to see certain kinds of pictures

00:06:22.260 | or you have to overpay them, right?

00:06:23.940 | And so what did we find?

00:06:25.060 | It was really, I think, scientifically revealing,

00:06:27.900 | but it's pretty fun.

00:06:28.820 | People got it immediately.

00:06:31.940 | They will pay--

00:06:33.180 | - Juice.

00:06:34.020 | - Juice.

00:06:34.840 | They will give up juice.

00:06:35.680 | They will pay it to see pictures of the perineum,

00:06:38.940 | the hind quarters of females.

00:06:41.180 | This was original study was in male monkeys.

00:06:43.240 | They will pay to see the faces of dominant males.

00:06:47.700 | And you had to pay them to see the faces

00:06:50.180 | of subordinate males.

00:06:51.980 | Okay, so females will give up juice

00:06:56.140 | to see the taints of testosterone rich male monkeys

00:07:02.020 | and male monkeys will pay juice to see the swollen taints

00:07:07.020 | of female monkeys that are, because of the swelling,

00:07:13.260 | indicates a better reproductive competence.

00:07:16.500 | - Yes, better, you know, douse the time.

00:07:18.820 | The time is ripe, okay, to mate.

00:07:21.740 | But it's just in general, it's a signal that is like,

00:07:24.220 | what we would say is it's important.

00:07:25.620 | It has value.

00:07:26.780 | - Monkey porn.

00:07:27.620 | - It's something you should track.

00:07:29.060 | And in fact, yeah, they're paying for it.

00:07:30.660 | So, you know, it just blew up on the internet.

00:07:32.620 | Even back then, it was like suddenly million,

00:07:34.580 | every website was like, oh, you've proven monkey porn,

00:07:36.740 | blah, blah, blah.

00:07:37.580 | It was kind of a fun ride.

00:07:39.860 | It was a New York Times idea of the year in 2005,

00:07:44.780 | which was, again, kind of shocking.

00:07:47.900 | You know, there's like, a little word on that.

00:07:50.580 | But people, it makes sense.

00:07:52.340 | And the thing I want to point out is that

00:07:54.340 | we ran this same experiment in people,

00:07:57.820 | not with unclothed humans.

00:08:01.260 | So we used, and we used only, well, no, it was,

00:08:05.060 | and we had to create our own stimulus set

00:08:06.980 | because all the stimulus sets that were out there

00:08:09.340 | for visual studies of humans were like a bunch of,

00:08:13.500 | you know, German people looking very dour.

00:08:16.460 | They were very well controlled.

00:08:17.580 | And we wanted something that was more natural.

00:08:18.820 | So we downloaded thousands of photos

00:08:21.700 | from this website, hotornot.com.

00:08:23.700 | I don't know if you recall that,

00:08:25.100 | but it was a website where you could upload pictures

00:08:27.020 | and people would rate you.

00:08:28.020 | I mean, now that's like-

00:08:30.940 | - Probably wouldn't be allowed now.

00:08:31.860 | I remember Rate My Pet.

00:08:34.780 | - Rate My Pet, Rate My Professor,

00:08:36.340 | I think, which is still around.

00:08:37.540 | - And we were saying rate.

00:08:38.860 | - Rate.

00:08:39.700 | - Rate. - Rate.

00:08:40.940 | - With a T, my pet.

00:08:42.980 | - Yeah.

00:08:43.820 | - But this was hotornot.com.

00:08:45.180 | So you get all these really natural looking.

00:08:47.540 | And then we had, this was really funny though too.

00:08:50.380 | So we had a group of,

00:08:54.180 | separate groups of raters

00:08:56.120 | from the people who we actually tested in the experiment.

00:08:58.740 | So we had, you know, a group of males,

00:09:01.420 | heterosexual males rating the female photos and vice versa.

00:09:05.140 | And that was interesting in its own right.

00:09:07.340 | So we were just trying to establish like,

00:09:09.620 | we're not saying why they're attracted or anything like that,

00:09:11.780 | just like, let's measure it, okay.

00:09:14.300 | And it was really fun because, you know,

00:09:16.100 | by the, and it took, it was hard work.

00:09:17.700 | You're having to do one every three seconds.

00:09:19.620 | And it took like an hour.

00:09:20.940 | And the, you know, when the women were done rating,

00:09:24.340 | they're like, whew, okay, I'm glad that's over.

00:09:26.880 | The hour's over and our male raters were like,

00:09:30.440 | did you have any more?

00:09:31.680 | You know, can I, I'd be happy to sit here

00:09:34.120 | and rate more photographs for you.

00:09:36.480 | - Interesting.

00:09:37.400 | So women got sort of like,

00:09:40.180 | they got tired of rating males for attractiveness.

00:09:43.560 | - Yes.

00:09:44.400 | - Males did not tire of rating females for attractiveness.

00:09:46.360 | - They did not at all, which is, that's anecdotal,

00:09:48.960 | but it's still, I think it's revealing.

00:09:51.200 | Then we ran the pay-per-view experiment,

00:09:53.960 | just like in monkeys, on humans.

00:09:56.800 | - Pay-per-view.

00:09:57.640 | - And we also ran a couple of other economic lead,

00:10:01.960 | you know, standard economic tasks.

00:10:03.520 | One would be, how long are you willing to wait?

00:10:05.480 | So that's a delay discounting.

00:10:07.080 | Like, in general, you will wait longer for a bigger reward.

00:10:11.000 | A smaller reward.

00:10:12.040 | And also how hard would you work?

00:10:13.520 | And we, the work was like,

00:10:14.960 | you had to alternate pressing two keys on a keyboard.

00:10:17.080 | It was really just menial, laborious, you know, et cetera.

00:10:21.400 | So, the two interesting,

00:10:24.140 | just sociologically it's interesting,

00:10:25.840 | what comes out of this.

00:10:27.000 | Our female subjects basically wouldn't give up money.

00:10:32.880 | They were working for money.

00:10:36.000 | They were hearing the sound of coins

00:10:37.260 | coming out of a slot machine,

00:10:38.360 | which was proportional to how much money they actually got.

00:10:40.760 | - Real money.

00:10:41.600 | - Real money.

00:10:42.420 | If you ignored the pictures,

00:10:43.440 | you'd go home with like $17 extra,

00:10:46.560 | compared to if you were influenced by them.

00:10:48.920 | And the females did really well economically.

00:10:51.280 | So they pretty much kind of ignored the pictures

00:10:53.520 | of the males, even though they were rated,

00:10:55.760 | even the ones that were super hot,

00:10:58.200 | they were not very concerned with that.

00:11:01.700 | For the males, it was exact opposite.

00:11:03.520 | So the males are giving up essentially.

00:11:05.440 | They're paying, and they had thousands of trials.

00:11:08.100 | They're paying somewhere between a half

00:11:09.800 | and three quarters of a cent to see images

00:11:13.800 | of women who were rated in the top,

00:11:15.680 | like third of attractiveness.

00:11:18.360 | They also would wait significantly longer

00:11:20.200 | and they would work really hard.

00:11:21.480 | It was like rats pressing for cocaine,

00:11:23.600 | quite literally to keep those pictures up on the screen.

00:11:27.560 | Okay, so that's the setting we've established in monkeys

00:11:31.640 | and in people, similar economic principles

00:11:34.240 | that are guiding social, you'd call it attention,

00:11:37.820 | social valuation, whatever.

00:11:39.440 | So we're like, okay, let's go look in the brain.

00:11:41.880 | So we did an MRI experiment, fMRI experiment,

00:11:44.560 | measured blood flow to different parts of the brain.

00:11:49.560 | We only tested males because they were the ones

00:11:54.560 | who displayed differential preferences there.

00:11:58.560 | And what we found is that kind of parts

00:12:01.360 | of the visual system that are involved in encoding faces,

00:12:03.920 | but then the reward system was activated

00:12:08.240 | and tracked linearly how much money

00:12:12.040 | these guys were paying to see images.

00:12:15.280 | There's basically the trade-off value, the currency,

00:12:18.320 | the translation of pictures into money, okay?

00:12:22.440 | Then in monkeys, we studied all the same areas,

00:12:25.180 | but now we could record from individual neurons

00:12:27.580 | in those areas, rather than looking at blood flow,

00:12:29.320 | which is a crude proxy.

00:12:31.520 | And we found exactly the same thing,

00:12:34.360 | which is that neurons in the reward system

00:12:37.520 | were spontaneously and strongly activated by those pictures,

00:12:42.880 | you know, that made sense, right?

00:12:46.680 | So the pictures of the perinea of females

00:12:49.200 | by dominant male faces.

00:12:52.500 | And that correspondence, I thought, was pretty compelling.

00:12:58.400 | (upbeat music)

00:13:02.780 | (upbeat music)

00:13:05.360 | (upbeat music)

Why Men & Women Show Attraction Differently | Dr. Michael Platt & Dr. Andrew Huberman

Chapters