Why We Cry & the Evolutionary Purpose of Tears | Dr. Noam Sobel & Dr. Andrew Huberman

00:00:00.000 | We started thinking about tears and looking into tears because tears are bodily liquid

00:00:08.560 | emotional tears that we emit in emotional situations where these are situations where

00:00:14.840 | nonverbal communication is critical and key.

00:00:20.000 | And tears are a liquid that is puzzling beyond ocular maintenance.

00:00:30.240 | And so the most influential text, I think, till this day in emotion research is Darwin's

00:00:39.040 | book The Showing of the Emotions in Man and Animals, I think is the full name of the book.

00:00:44.960 | And an entire chapter, chapter 6, is devoted to tears, an entire chapter of this book.

00:00:51.760 | Why?

00:00:52.760 | With no conclusion.

00:00:53.760 | Why?

00:00:54.760 | Because the book revolves around describing the functional antecedents of emotional expressions.

00:01:03.560 | So for example, showing of the teeth is a sign of aggression, right?

00:01:07.880 | So animals first bit with their teeth and Darwin argued that through evolution just

00:01:13.520 | showing the teeth alone became an aggressive sign because it started from biting.

00:01:18.960 | Or what I find is a beautiful example, and this is work partly done by Adam Anderson

00:01:23.360 | now at Cornell, is the emotional expression of disgust.

00:01:30.280 | So disgust, which comes from the Latin disgusia, distaste, right, is spitting something out

00:01:36.720 | of your mouth.

00:01:38.040 | Now what Adam showed is that the musculature patterns of activation and the temporal sequence

00:01:46.000 | of activation when you experience moral disgust are the same as when you spit a bitter taste

00:01:52.400 | out of your mouth, right?

00:01:53.400 | So again, so there's a functional antecedent spitting something out.

00:01:58.280 | And through evolution, the argument was that it became an expression of emotion and you

00:02:02.280 | express disgust just as if you're spitting something out of your mouth, even though in

00:02:06.320 | the case of moral disgust, there's nothing you're spitting out of your mouth.

00:02:10.380 | So Darwin systematically went through the expressions of emotions and for each one went

00:02:14.520 | to their functional antecedent and explained everything very nicely.

00:02:18.200 | And then he got stuck with tears, right?

00:02:19.820 | Because tears are an obviously emotional expression and he could not find a functional antecedent.

00:02:27.560 | So he ended up saying this is an epic phenomenon, basically, right?

00:02:31.560 | What all scientists do when they don't have a good explanation, blame it on nature.

00:02:37.040 | Right, right.

00:02:38.040 | So, but he bothered to write this entire chapter on the ocular sort of maintenance function

00:02:44.280 | of tears and so on and so forth, but nothing emotional.

00:02:47.320 | So we thought, well, maybe the function is a chemical signal.

00:02:52.440 | And so with that in mind, we harvested emotional tears, which was also an amusing event on

00:03:01.760 | its own, right?

00:03:02.760 | Because we posted messages on all sorts of boards that were seeking experiment participants

00:03:16.160 | who cry with ease.

00:03:19.000 | Now this generated an unfortunate gender bias in our study, right?

00:03:22.920 | Because we received about a hundred women volunteers and about one man.

00:03:27.160 | And you know, I think this is not a problem only in macho Israel, right?

00:03:30.620 | Probably anywhere in the West.

00:03:31.620 | This would be, I mean, definitely in America would be the same, I think.

00:03:34.000 | My guess is that there are probably men out there who cry easily, emotional tears easily.

00:03:38.520 | Oh, I'm sure.

00:03:39.520 | But they're not going to show up.

00:03:40.520 | Yeah, that's what I'm saying.

00:03:41.520 | It's a cultural thing.

00:03:42.520 | It's not, you know, you're not going to come to a lab and say, you know, I cry all the

00:03:44.400 | time.

00:03:45.400 | It's just not going to happen.

00:03:48.280 | And then what we did is for each one of these participants, you know, we would ask them,

00:03:54.920 | you know, is there a particular film event that you know of that, you know, a scene that

00:03:59.000 | makes you cry?

00:04:00.000 | And interestingly, in these effective criers, there's always, oh yes, you know, the scene

00:04:05.160 | in so-and-so, I always cry profusely from that.

00:04:09.000 | You know, they have their-

00:04:10.000 | Can you give me an example of one of the more commonly named scenes?

00:04:13.200 | With these, the movie The Champ.

00:04:18.240 | The Champ dies, he's a boxer.

00:04:21.320 | And he dies and literally in the hands of his about eight-year-old son.

00:04:28.640 | And his son is standing next to his bed and, you know, saying champ, champ.

00:04:35.600 | And he dies, right?

00:04:38.660 | It's a winner.

00:04:39.660 | Okay.

00:04:40.660 | Waterfall.

00:04:41.660 | Yeah, yeah.

00:04:42.660 | Got it.

00:04:43.660 | So, you know, we're probably the neurobiology lab with most sad movie films on the shelf

00:04:49.000 | in the world, right?

00:04:50.000 | We have a whole huge collection.

00:04:51.000 | There is such a thing as tears of joy, by the way.

00:04:54.200 | So, no.

00:04:55.200 | No?

00:04:56.200 | Well, we're going ahead of ourselves.

00:04:57.200 | Like I said, we tried to collect them and failed.

00:05:00.560 | Even people who think they shed tears of joy and laughter, their eyes water a bit.

00:05:04.940 | But it's not the same thing.

00:05:08.400 | In the effective criers we end up screening, so we collect a full ML of tears, a full ML

00:05:16.560 | of tears in about 15 minutes.

00:05:19.000 | So that's pouring, right?

00:05:21.080 | And that doesn't happen from laughter.

00:05:23.800 | Or we've never seen that.

00:05:25.000 | We've never seen that happen from laughter.

00:05:28.800 | We tried.

00:05:30.760 | So we have all these sad films.

00:05:33.040 | And by the way, one of the amusing things is when we ultimately published this paper

00:05:37.640 | in Science, we were forced, in retrospect, to go out and actually buy the films.

00:05:44.760 | Right?

00:05:45.760 | I mean, you know, originally we like downloaded them, but you can't because you'd be violating

00:05:50.440 | copyright laws, right?

00:05:54.500 | So we had to buy, like purchase all these films that the participants watched.

00:05:59.240 | So we actually have these in lab, like DVDs, you know, that we actually purchased.

00:06:04.240 | Nice coverage of potential legal fallout there, Noam.

00:06:09.040 | No, we did.

00:06:10.040 | We did.

00:06:11.040 | I believe you.

00:06:12.040 | I believe you.

00:06:13.040 | So, yeah.

00:06:14.040 | And, well, we can touch on that later.

00:06:19.160 | But so most of these volunteers who come saying they can cry with ease actually don't meet

00:06:29.560 | the bill.

00:06:31.200 | And so out of the about hundred, at least, more women that we screened, we ended up with

00:06:36.320 | about six who could really come to lab week after week in poor tears.

00:06:44.040 | There's a name for this in psychiatry.

00:06:45.540 | They call it a narrative distancing.

00:06:48.720 | Some people, when they watch a film where someone's getting hit, they flinch quite a

00:06:54.480 | lot.

00:06:55.480 | It's almost as if they're experiencing it.

00:06:56.880 | But it works in the opposite direction, too.

00:06:58.400 | I know someone like this, where if they watch a film that someone's experiencing something

00:07:03.320 | even mildly positive, their mood elevates.

00:07:05.760 | So they can quickly bridge.

00:07:09.100 | And it's not always adaptive, as you can imagine.

00:07:11.240 | So there's a lack of narrative distancing.

00:07:13.080 | Right.

00:07:14.080 | Yeah.

00:07:15.080 | Well, one issue you can bring up with this entire line of studies in our lab is I don't

00:07:17.760 | know if there's something very unique about the donors, right?

00:07:21.680 | I mean, we're assuming these are tears.

00:07:23.160 | No, this is pretty common.

00:07:24.400 | I think that the numbers I saw out there, about 5% to 8% of people got about 600.

00:07:33.320 | So we collected tears, and we exposed participants to these tears.

00:07:44.880 | And we found a few things.

00:07:46.720 | First of all, the tears are completely odorless.

00:07:49.260 | You cannot detect them at all.

00:07:51.840 | Completely odorless.

00:07:54.960 | And yet, when you sniff them, you have a pronounced reduction in testosterone within about 20

00:08:05.400 | minutes, half an hour.

00:08:06.880 | This is men and women smelling women's tears.

00:08:09.560 | Just men.

00:08:10.560 | Men smelling women's tears, but not perceiving any odor.

00:08:13.280 | Nothing.

00:08:14.280 | Just sniffing them.

00:08:15.640 | And you have about a 14% drop in free testosterone.

00:08:21.680 | Free.

00:08:22.680 | OK, so this is testosterone that's already been liberated from the testes.

00:08:26.080 | Free testosterone.

00:08:27.080 | We've done a few hormones that's either bound or unbound-- is unbound, excuse me-- from

00:08:32.560 | sex hormone binding globulin, et cetera.

00:08:34.320 | And it's the active form.

00:08:35.480 | So it's subject to very short timescale changes.

00:08:41.120 | Yeah.

00:08:42.120 | And this is-- people who study testosterone, which is not me, but they tell me this is

00:08:47.480 | a really strong effect.

00:08:49.400 | It's hard to even pharmacologically get an effect like that that fast.

00:08:53.320 | No one in pharmacology.

00:08:54.320 | Yeah.

00:08:55.320 | Years ago, I spent time studying endocrine effects of this sort.

00:08:57.560 | And that's a tremendously sized effect.

00:09:00.640 | So here, I'll point out in passing that one of the concerns we had because of the effort

00:09:09.800 | to run this study is that nobody would ever try to replicate it.

00:09:14.800 | And to our joy, about two years later, an independent group from South Korea, OHL, who

00:09:22.320 | I don't know at all, replicated the testosterone effect to a T. I mean, like, same numbers.

00:09:29.840 | So it lowers testosterone.

00:09:34.480 | And we then also looked, using MR, at the effect on brain activity and saw a pronounced

00:09:47.800 | effect on activity, a dampening, a lowering of activity under an arousing state, a lowering

00:09:54.560 | of activity both in the hypothalamus and in the fusiform gyrus, for whatever reason I

00:10:01.480 | don't know.

00:10:02.480 | Face recognition area.

00:10:03.480 | It's other things, yes.

00:10:05.800 | And we don't know why, but pronounced.

00:10:10.240 | And currently, Shania Groen in our lab is replicating this again, and this time with

00:10:18.160 | a stronger behavioral component.

00:10:21.400 | And I can share with you unpublished data now under review that's, as you would expect,

00:10:28.520 | given the effect on testosterone perhaps, sniffing tears lowers aggression in men.

00:10:34.600 | Using again the TAP, the same experiment used by Yvonne in the hexadecanol experiment.

00:10:39.040 | The TAP, I'm going to think of that as the SATIS, the titration, the SATIS titration

00:10:45.880 | experiment.

00:10:46.880 | The Tyler Aggression Paradigm.

00:10:47.880 | Not unlike the Milgram experiments of the 1950s, which post, this was looking at sort

00:10:55.160 | of post-Holocaust behavior, people basically in American laboratories thinking they were

00:11:01.720 | torturing other people simply because they were told to.

00:11:04.800 | And a lot of people did that, even though most people would report that they would never

00:11:07.960 | torture somebody else.

00:11:08.960 | Yeah, no, humans are not a wonderful species.

00:11:11.720 | Or as you could say, I think it was the great Carl Jung that said, we have all things inside

00:11:16.960 | of us, but the goal is not to experience them all, certainly.

00:11:23.920 | It's an incredible study and it points again to the power of these chemosensory systems

00:11:31.080 | and pathways.

00:11:32.080 | And obviously, there's so much here.

00:11:35.320 | I don't know if you want me to tell about this or not, and I guess you can edit it out.

00:11:40.800 | This is just sharing stories about the politics of science.

00:11:46.840 | So whereas the effect on testosterone was replicated by an independent group.

00:11:53.840 | In the original study in science, it had three components.

00:11:58.480 | One was the effect on testosterone, which was robust.

00:12:02.360 | The second, which was brain activity, which was robust.

00:12:06.320 | And there was a significant but weaker effect on behavior.

00:12:11.180 | And I don't think we studied the right behavior in retrospect.

00:12:13.720 | What we looked at then was ratings of arousal associated with pictures.

00:12:21.240 | And there was an effect, it was significant, but it was not what carried the story.

00:12:28.640 | Now there is a lab in Holland of a guy by the name of, I'm probably mispronouncing this,

00:12:38.480 | but I think it's Wingerhoets.

00:12:42.240 | For the non-Dutch.

00:12:43.240 | Yeah.

00:12:44.240 | Dutch names are always a little bit of a challenge.

00:12:46.800 | And I shouldn't say that, being an Israeli, I shouldn't go too much on that line.

00:12:51.280 | But that lab really didn't like our original tear story.

00:12:58.640 | And the reason they didn't like it is because they've built a career on this notion, including

00:13:06.960 | a book with this title, that emotional tears are uniquely human.

00:13:12.440 | Now, here I should, well, I should share.

00:13:17.240 | So one of the things we really liked about the tear result is that partially before we

00:13:25.720 | did our work, but more afterwards, and we like that because usually things, so usually

00:13:29.600 | in our chemo signaling work, like what I told you before about the Bruce effect, we look

00:13:33.000 | at what happens in rodents and we see if the same thing is happening in humans.

00:13:36.600 | This was a rare case where after we did this work, more or less identical effects were

00:13:42.680 | discovered in rodents.

00:13:44.400 | So a paper published in Nature two years later found that mouse tears, mouse pup tears, lower

00:13:52.000 | aggression in male adult mice towards them.

00:13:56.680 | In a smell dependent way.

00:13:58.600 | Yeah.

00:13:59.600 | Yeah.

00:14:00.600 | So and they also actually found the actual component in tears that, so the tear pheromone

00:14:05.200 | that lowers aggression.

00:14:06.600 | Right.

00:14:07.600 | So, you know, this has us thinking of tears as you can think of tears as like a chemical

00:14:12.000 | blanket in a way that you're covering yourself up again with, you know, to protect against

00:14:16.800 | aggression, right?

00:14:20.560 | And so our finding, you know, which to me, I mean, this is consistent with how I think

00:14:25.120 | about behavior in general.

00:14:26.120 | I, you know, I don't think, you know, beyond language, there are very few things, definitely

00:14:30.960 | sensory things that are uniquely human.

00:14:33.440 | You know, I'd be hard pressed.

00:14:36.320 | But so, you know, our finding went against their story, right?

00:14:41.480 | Because, you know, here we're saying no, you know, tears are this chemo signaling mechanism

00:14:45.340 | like all animals.

00:14:46.960 | And by the way, you know, just after this entire debate, about six months ago, there

00:14:51.840 | was a paper in Current Biology that dogs emit emotional tears.

00:14:56.800 | And it was the dogs emit emotional tears when they reunite with their owners.

00:15:02.040 | And you were talking before about oxytocin.

00:15:06.640 | So I think what they showed there is that not only that, but that the view of seeing

00:15:13.520 | the tears in the dog influences oxytocin in the humans.

00:15:17.880 | I hope I'm getting this right.

00:15:20.120 | No, I absolutely believe this.

00:15:21.880 | I mean, from the time I brought Costello home at eight weeks old.

00:15:26.240 | Costello is your dog.

00:15:27.240 | He's my dog.

00:15:28.240 | Unfortunately, he passed away.

00:15:29.240 | We haven't had him in a long time.

00:15:30.240 | But from the time I can recall crying, listen, I've certainly cried before many times in

00:15:36.300 | my life, many, many times.

00:15:39.960 | The only time I ever recall crying to the point where I wasn't sure that I could keep

00:15:44.000 | producing tears, but somehow it is when I had to put him down, right?

00:15:47.540 | Is this like, you know, and if I talk about too long now, I'll start crying.

00:15:50.880 | You know, it's one of those things.

00:15:52.360 | I think it's a healthy emotional state.

00:15:55.000 | But I recall when he was a puppy, thinking this oxytocin thing must be real.

00:16:00.160 | Because I can recall being in faculty meetings, which, you know, fairly stated are not always

00:16:05.880 | that interesting, but they can be pretty interesting.

00:16:07.720 | And someone presenting data and my mind thinking, I hope Costello is okay.

00:16:12.680 | What's he doing down in my office?

00:16:13.680 | This is when he was very little.

00:16:15.360 | And also not needing to eat, not being able to focus on anything else except my attachment

00:16:21.520 | to him for about the first two or three weeks that I had him.

00:16:24.080 | Then it was easy.

00:16:25.080 | Then I could focus off on other things.

00:16:26.520 | And I think that dogs, perhaps through oxytocin, hijack the circuitry that's intended for child

00:16:32.840 | rearing.

00:16:33.840 | I really do.

00:16:34.840 | Otherwise, why would people be so ridiculously attached to their dogs?

00:16:38.160 | Hence all the posts of everyone thinks their dog is the cutest dog, the same way everyone

00:16:41.800 | thinks their children are the cutest children.

00:16:43.800 | Costello, by the way, was a very handsome bulldog.

00:16:50.160 | So again, so even, you know, to put another nail in that story of tears are uniquely human.

00:16:56.480 | So they're not.

00:16:57.600 | Dogs shed emotional tears.

00:17:01.360 | And so that group really didn't like this.

00:17:04.800 | And they went ahead and tried to replicate, and to your listeners I'm showing double quotations

00:17:12.000 | on the replicate, only the behavioral part, the ratings of arousal of women and failed

00:17:23.120 | to replicate that.

00:17:25.240 | Now, this was just sharing on how science works and doesn't work in my notion in this

00:17:30.880 | case.

00:17:32.120 | So at the time, after they got this accepted in some journal, not a field journal, a journal

00:17:43.240 | of memory of something, they contacted me for a response.

00:17:52.280 | And I wrote to the office and I said, look, you know, this is very odd to me, why don't

00:17:57.440 | you come, why don't we replicate this again together and see if it doesn't work.

00:18:01.920 | If it doesn't work, I'll publish it with you that it doesn't work, but you know.

00:18:06.760 | And so I said, why don't you send over a graduate student or the lead author and we'll do it

00:18:10.240 | here and we'll show them how it's done because they did it very wrongly in the paper.

00:18:16.120 | And so they replied that no, they don't have money to send over a graduate student to do

00:18:20.160 | it.

00:18:21.160 | So they replied saying, okay, I'll fund the graduate student coming over and I'll fund

00:18:25.800 | the entire study and their stay and so on and so forth and let's do this together.

00:18:30.240 | And they replied, no, they're not willing to do that, which I don't think is the way

00:18:34.980 | things should work.

00:18:37.000 | And they published this sort of failed behavioral effect in that paper.

00:18:44.080 | So I'm just sharing this, you know, that it's not only, there was that successful replication

00:18:48.320 | with the effect on testosterone, but there was supposedly this failed replication on

00:18:52.040 | the effect in behavior.

00:18:55.560 | And then I published a rebuttal on that, which I don't know if I should have done, but I

00:18:59.960 | did.

00:19:00.960 | Well, I think it's interesting.

00:19:01.960 | I mean, I think provided studies are done correctly.

00:19:05.720 | I mean, the positive result almost always trumps the negative result.

00:19:10.560 | And yet I think replication is key.

00:19:12.120 | The problem, as you pointed out, is that replication is rarely pure replication of the exact study.

00:19:16.760 | This one is not even remotely, but I published the detail.

00:19:20.240 | So actually they hid something in their data that did partially work.

00:19:23.880 | So I asked for their data and I reanalyzed it and that's what I published in the rebuttal.

00:19:28.040 | But you know, this is just sharing on how science works.

00:19:30.760 | I took advice.

00:19:31.760 | So it's not that I'm friends with him, but at that time I was communicating a bit because

00:19:37.320 | we were on some board with Daniel Kahneman, who's Nobel laureate.

00:19:43.880 | And so I asked him, how should I deal with this?

00:19:48.860 | You know, give me some advice here.

00:19:50.180 | I was really, you know, it was emotionally not fun to be in that position.

00:19:55.920 | And he said, "Don't ever publish a rebuttal.

00:20:01.160 | Don't do anything."

00:20:02.160 | And I was, you know, how can I?

00:20:04.280 | You know, I have to do something.

00:20:05.280 | He said, "No, don't.

00:20:06.400 | Because once you do that, then, you know, people don't go into the details.

00:20:10.080 | They won't read the details of your rebuttal.

00:20:11.640 | They'll be like, well, there's a group that says this and there's a group that says that.

00:20:15.000 | So it's unclear."

00:20:16.000 | Well, and yeah, I mean, I appreciate that you're bringing it up today.

00:20:19.280 | And I do appreciate that you published the rebuttal and that you offered in a very magnanimous

00:20:24.040 | way to do a collaboration.

00:20:25.360 | That's what he then said.

00:20:27.160 | So Kahneman's advice after that was that, well, if you insist, then just publish, write

00:20:34.000 | a response that you offered them to come do it together.

00:20:37.080 | They refused and there's nothing you can do about that.

Why We Cry & the Evolutionary Purpose of Tears | Dr. Noam Sobel & Dr. Andrew Huberman

Chapters