JAD: This is Radiolab. I'm Jad.

ROBERT: And I'm Robert.

JAD: And today's program is about sleep.

ROBERT: As in the kind of sleep where you perchance to dream.

JAD: Exactly. Did you know that story about the benzene molecule?

ROBERT: No.

JAD: Speaking of dreams.

ROBERT: No.

JAD: Well, here ...

ROBERT: The benzene mol ...

JAD: 1865. German chemist is trying to figure out the shape of this molecule, benzene. He knows it has a certain amount of one kind of atom and a certain amount of another, but it can't figure out how they all link up.

ROBERT: Right?

JAD: And he's tortured by this problem. Goes to sleep.

ROBERT: Yeah?

JAD: Has a dream of a snake biting its tail. Wakes up. Bolts right up, and says, "It's a ring. It's a ring!"

ROBERT: [laughs] Do you believe that?

JAD: I want to.

ROBERT: Well, yeah.

JAD: I mean, don't get me wrong. I hate it when people tell me their dreams. Hate it. I want to stab my eye with a fork, frankly, when people tell me their dreams. I don't know why. It just drives me ...

ROBERT: Well, I'm never gonna tell you about my dreams again.

JAD: Good.

ROBERT: [laughs] But, you know, you're not alone, because for a long time scientists have avoided studying dreams, because they think they're so random and meaningless and unstudyable.

JAD: Right.

ROBERT: But we did meet a guy ...

BOB STICKGOLD: I'm Bob Stickgold. S-T-I-C-K-G-O-L-D. I'm an associate professor of psychiatry at Harvard Medical School.

ROBERT: ... who found an interesting way to ask the question ...

BOB STICKGOLD: Why do we dream? Simple question, very hard answer.

ROBERT: Robert Stickgold was one of the first modern scientists to take dreams seriously. And for him, it actually began kind of by accident.

BOB STICKGOLD: I had been up in Vermont with my family. We had gone and climbed Camel's Hump, one of the higher easy-to-climb mountains in Vermont. We'd gone at eight in the morning, we were back at two in the afternoon.

ROBERT: And for that whole day, he'd been up climbing on the rocks, gripping them with his hands, really climbing. Later that night ...

BOB STICKGOLD: I lie down. I close my eyes. I can feel the rocks under my hands. And I sort of startle up and I say, "Whoa, that's really bizarre!" It wasn't like I was thinking about it. I was there. I could feel the rock. I'd been off the mountain for eight hours. Nothing like that had happened. I lie down in bed for three minutes, starting to go to sleep and boom, it's there. And I tried again and I fell asleep. Two hours later, I wake up, have to go to the bathroom. I go to the bathroom. I come back, and I say, "That was way cool. I have to try that again." And I cannot get it back. What happened in those two hours to those memories that they won't intrude anymore?

BOB STICKGOLD: And then I started talking to friends, and they say, "Oh, try canoeing." Or someone else says "Try—try whitewater rafting if you want to get that." And someone else says, "Oh, hello! Take organic chemistry. And you go to bed at night, and all you see are these bloody molecules rotating in front of your eyes." Those daytime activities are affecting your dreams.

ROBERT: And that got him thinking: what exactly is the connection between what you do during the day, and what you dream at night.

JAD: What are the rules of that? He figured all right, well this replay is kind of interesting. Maybe I'll test it. But how? If I get some subjects together, what could I have them do during the day that would reliably end up in their dream?

ROBERT: Well, you can't have them all go for a hike.

JAD: Mm-mm.

BOB STICKGOLD: And I'm probably not gonna get permission to take them whitewater rafting.

JAD: Too expensive.

ROBERT: So what could he do?

BOB STICKGOLD: Sort of sat fallow for a year. And I was moaning to some of my students about how I can't think of how to do this, and someone says "Tetris." And somebody else says, "Absolutely." And I said, "What? What?" They said, "Well, don't you play Tetris?" And I say yes. Well, when you start playing Tetris, turns out when you start playing Tetris, you go to bed at night, you lie down in bed and you see Tetris pieces falling down in front of your eyes.

ROBERT: Sure.

JAD: Oh, yeah.

BOB STICKGOLD: You knew that?

JAD: Absolutely.

ROBERT: In front of your eyeballs.

BOB STICKGOLD: You guys both know that?

JAD: Oh, yeah.

BOB STICKGOLD: I got a cover of Science magazine for the first published paper on dreaming in 40 years because I discovered that, and everybody already knew it!

ROBERT: [laughs]

ROBERT: It was that simple. He got a bunch of people, put them in a room, had them play Tetris. Later that night, they woke up, and 60 percent of them were dreaming of Tetris. 60 percent!

JAD: How do you know that? I mean. just from their reports?

BOB STICKGOLD: They report. As they're falling asleep, we're monitoring them, electrophysiologically. And as they start to drift off to sleep, "Please report now."

JAD: This Tetris observation was a pretty good start in terms of getting at that question.

BOB STICKGOLD: Why do we dream?

JAD: Why do we dream? How does it work? What if, as a next step, instead of having the people report their own dreams, waking them up and doing that whole thing, what if instead, you could cut the person out of the equation entirely and go right to the source, to the dream directly?

MATT WILSON: Matt Wilson. I'm a researcher here at MIT, and I'm a neuroscientist studying learning and memory.

JAD: That's what Matt Wilson does. He takes us to the Dream Lab.

MATT WILSON: So when we first come in, what we see is this bank of monitors.

JAD: 13 monitors, all in a row.

MATT WILSON: Each monitor displaying ongoing activity in the brain, with little panels, each panel showing these ...

JAD: It's like the Kennedy Space Center, really. All the monitors have data just flashing all over them—graphs and squiggly lines and numbers. It's not immediately clear where all this information is coming from, but if you peek around the back, you'll see that all the computer wires go to one box, which then connects to a cable, which then goes up to the ceiling, over to a wall, and down into the head—into the head!—of one tiny rat.

MATT WILSON: Here he is.

JAD: He's just kind of hanging out in his own little basket.

MATT WILSON: You see him just resting.

JAD: Is that the little guy himself?

MATT WILSON: Yeah, that's him.

JAD: He looks pretty normal, except for this cable coming out of his skull. And the cable is basically a microphone, or a bunch of them, which Matt uses to eavesdrop on the brain cells inside the rat's head as they chitchat. And this is what that sounds like ...

[scratching, popping sounds]

MATT WILSON: You can hear this kind of snap-crackle-pop sound. These are individual cells that are firing.

JAD: Like right there. One of those.

MATT WILSON: And that kind of whooshing sound ...

JAD: Uh-huh?

MATT WILSON: I can tell this animal's sitting, resting quietly.

JAD: Amazingly, he says this while he has his back to the animal. He is so fluent with the Morse code language of the rat's brain cells, he doesn't even have to actually look at the animal to know what it's doing. He can just instantly decode all of that snapping.

ROBERT: Hmm.

JAD: Kind of like that guy in The Matrix, the bald guy.

[ARCHIVE CLIP, The Matrix: I don't even see the code. All I see is blonde, brunette, redhead.]

JAD: Just by listening, Matt knows when the animal is sitting, he knows when it is sleeping, he knows when it's running around in a maze, he even can tell which direction it's running.

MATT WILSON: It just happened that as we were studying these patterns while the animal ran around, after the experiments, the animals would—they would get tired, they would go to sleep. I would be there in the room, but I would continue to listen to the activity.

[ARCHIVE CLIP, Matt Wilson: Notice how it's gotten silent?]

[ARCHIVE CLIP, Jad: Yeah.]

MATT WILSON: I began to notice that when the animals were asleep, the brain cells weren't just firing randomly. In fact, when animals would go into REM sleep ...

[ARCHIVE CLIP, Matt Wilson: So now he is, in fact, going into REM right now.]

MATT WILSON: ... the pattern of activity that you could hear ...

[ARCHIVE CLIP, Matt Wilson: Notice that it's not these whooshes anymore.]

MATT WILSON: ... sounded very much like the pattern that the animal had just been running through. In fact, if you weren't watching the animal, you would think, "Oh, the animal has gotten up and is running around again." But then you turn and you look and you see the animal is asleep.

JAD: He checked the data, and it wasn't simply that the rat was running around in its mind while its body was asleep, it seemed to be running a specific route—the same route, in fact, that it had run earlier in the day.

ROBERT: The same sequence, same order, same everything?

MATT WILSON: Yes.

JAD: It was rerunning its maze step for step. So then he asked the next question.

MATT WILSON: Are they seeing the things that they saw while they were awake? We can actually look into these questions as a rat.

ROBERT: And? [laughs]

MATT WILSON: So the answer is we see evidence of replay in basically all of the parts of the brain that we have looked in.

JAD: They see the maze that they ran through? The very same maze?

MATT WILSON: Yes, they see the maze.

JAD: So that is dreaming, in a sense.

MATT WILSON: Well ...

JAD: They are ...

MATT WILSON: ... how do we define dreaming? So ...

JAD: Sounds like dreaming to me. I mean, I don't know. But the question remains: why would the rat or any creature do this? And so Matt came up with a simple next experiment. He decided to give the rat two mazes.

ROBERT: What would that do to its dreams of the night, or whatever you want to call them?

MATT WILSON: If they run on maze number one and then on maze number two, we see them running maze one and maze two together in a way that they did not experience when they were awake.

ROBERT: Oh!

JAD: Oh, so it's like a remix.

MATT WILSON: Exactly.

JAD: A new pattern that includes part of maze one and part of maze two.

MATT WILSON: Precisely.

JAD: Turns out that when the rat had more than one maze in its memory, it began to invent completely new mazes.

MATT WILSON: This gives us the thought that sleep is this unique opportunity to basically run through events, to put them together in ways that may not have occurred while the animals were awake. And that's what learning really is. Learning is about synthesis, about taking things that were apparently unrelated and figuring out the connection, that is figuring out the rules, the hidden rule, the undiscovered rules that will allow us to create something new.

BOB STICKGOLD: I think dreaming is a time when we try out possibilities that in waking we might not feel were worth trying. And when it really works, it can be profoundly important.

ROBERT: If Robert Stickgold is right, then how does this solving the problem thing, how does it work? How does the brain decide what to put into a dream and what to leave out of the dream?

BOB STICKGOLD: One of the interesting things about dreams is that people don't have dreams where they're word processing, where they're surfing the 'net. These things that they spent huge amounts of their day doing don't get into their dreams.

JAD: But somehow Tetris gets in there every time.

ROBERT: Every time!

JAD: And why would that be?

ROBERT: Well, he has a hunch.

JAD: Which he's actually exploring with a completely different video game.

BOB STICKGOLD: We've moved to a game called Alpine Racer, which we bought out of an arcade.

JAD: Which he showed us. Took us down the hall to the game room.

BOB STICKGOLD: Here we are.

ROBERT: And there in the corner, it stood.

JAD: Mockingly!

JAD: Oh, wow! It's a full-body game.

BOB STICKGOLD: Please step up.

ROBERT: I stepped up to the game. Got on the platform.

BOB STICKGOLD: It's still warming up.

ROBERT: And then I set off down a virtual mountain.

ROBERT: All right, I'm going downhill. I'm also a girl. I'm also avoiding the skis. Make a nice little turn there. Careful of the wall. Straight down ahead, and down we go. Oh, no the tree! [yells] We're not gonna go through the tunnel. This is a—ow! Oh, that hurt! [laughs]

ROBERT: As you can hear, this game was really stressful.

JAD: Yes.

ROBERT: Which is by design. Robert Stickgold has the theory that as you go through your day, your brain is constantly keeping track of emotions. That's the thing: emotional content. Like when you run into a virtual tree, for example, your brain is gonna flag that stuff.

BOB STICKGOLD: It's gonna flag that it's important.

ROBERT: It says, "Ooh, I need to remember this so I can work on it later. I'm gonna put a sticky on this one."

BOB STICKGOLD: So if it puts a sticky on everything that's hard during the day, then all the brain has to do when it's creating a dream is go and grab stickies.

ROBERT: Oh, and then I died! But I died nice.

BOB STICKGOLD: Just for the record, you got further than Jad on your first try.

JAD: Wow! Wow! It's like "Aaaah" and it's over. [laughs]

ROBERT: [laughs] Could you say that again?

JAD: Well, so you have people play Alpine Racer for 45-minute bursts throughout the day. What happens next? You wake them up?

BOB STICKGOLD: We monitor their brain activity. And just as they're falling asleep, within the first two minutes after they fall asleep, we'll wake them up.

[ARCHIVE CLIP, loudspeaker voice: Please report now.]

BOB STICKGOLD: There's a microphone right next to them on their bed, and they just report what was going through their minds.

[ARCHIVE CLIP, woman: I was just thinking about skiing.]

[ARCHIVE CLIP, woman: Skiing.]

[ARCHIVE CLIP, man: Skiing.]

BOB STICKGOLD: And we get on the first night up to 40 percent of all the reports being about skiing.

[ARCHIVE CLIP, woman: And the game that I've been playing.]

[ARCHIVE CLIP, man: Alpine Ski Racer, I guess.]

BOB STICKGOLD: 40 percent. Almost half of them. And that's right up there with what I would expect to see after trauma where something has been labeled so intensely that the brain says, "Okay, it's obvious what's on the agenda for tonight."

JAD: Stickgold thinks he's seeing the outline of the dream-making process here. It starts really simply at the very beginning of sleep. Like, right after you fall asleep?

ROBERT: Hmm.

JAD: With the replay. This, he suspects, is just the brain emptying out its stickies.

ROBERT: Things that really intrigued me during the day, that I felt during the day.

JAD: Yeah. But ...

BOB STICKGOLD: What happens if we let the people go to sleep, sleep two hours like I did in that very first time after climbing the mountain, wake them up after two hours of sleep?

ROBERT: Because remember, he couldn't get back the memory of the rocks after he'd spent two hours asleep.

JAD: That's right. And what he's found is that if you fast-forward two hours into the dream ...

BOB STICKGOLD: Get almost no reports of skiing at all.

JAD: ... the replay seems to dissolve into a remix.

BOB STICKGOLD: We start getting reports like, "Oh, I dreamt I was sliding down a hill."

[ARCHIVE CLIP, woman: Like I'm going downhill.]

BOB STICKGOLD: "Just rolling down a hill."

[ARCHIVE CLIP, woman: Downward motion.]

[ARCHIVE CLIP, man: I was thinking about—I was about to say downhill banana.]

[ARCHIVE CLIP, woman: I was thinking about skateboarding.]

[ARCHIVE CLIP, woman: I was thinking about ...]

[ARCHIVE CLIP, man: I was thinking about a bunch of bananas.]

[ARCHIVE CLIP, woman: Doing yoga on a ski slope.]

BOB STICKGOLD: Someone else had a dream that they were rushing through a forest with their body incredibly stiff and their legs not moving at all, as if they were on a conveyor belt.

JAD: It's like as the dream goes on, the brain is starting to free associate. "What do I have in my past that has anything to do with mountains, anything to do with crashing or skiing, anything at all that can help me?"

BOB STICKGOLD: What do I have in all my memories—in my case from the last 60 years—that fits associatively, thematically?

JAD: And the result? Well, it might seem random—it is—but every so often, he says, you come up with the right answer.

BOB STICKGOLD: So now we get to your dreams of people discovering the structure of benzene. Kekulé was his name.

JAD: Kekulé, actually, was his name.

ROBERT: [laughs]

JAD: August Kekulé.

ROBERT: That's close.

JAD: He was a German guy I talked about earlier who had a dream of a snake eating its tail, and realized from that dream that the shape of the benzene molecule is a ring. I don't know if that dream is true, but maybe that is, in fact, the point of dreaming. It's this time when you shut off the outside, turn inside, take the problems that you've got and start to really work on them. Pull them apart, make connections that you wouldn't normally make during the day.

ROBERT: However, have you ever wondered why it would be necessary when solving problems like this to dream so vividly?

JAD: Hmm.

ROBERT: Are you at all puzzled by the super-duper Technicolor, extraordinarily cinematic quality of some of these things? Because If it were just an everyday brain function to sort of make sense of the world and allow you to make new connections, you really wouldn't need quite the movie quality.

BOB STICKGOLD: So when we talk about dreams, what seems to come into dreams are memories, concepts, relationships, associations that have a strong emotional flavor. And I'm guessing from the data need a full-blown orchestration to be properly processed. And it is, it's Technicolor. The colors are overwhelming, almost.

JAD: So if I hear you right, what you're saying to Robert's question about why are the dreams so vivid is that I don't know, but maybe the vividness helps?

BOB STICKGOLD: That whole long answer is what a Harvard professor says instead of saying "I don't know."

JAD: [laughs]

ROBERT: [laughs]

JAD: I guess we should wrap up. If you want any more information on anything that you heard this hour, visit our website Radiolab.org. You'll see it there.

ROBERT: There'll be more information about sleep on there, won't there?

JAD: Yes, there will be more information on sleep. And you can sign up for our podcast: Radiolab.org. Also while you're at it, send us an email: Radiolab(@)wnyc.org is the address.

ROBERT: Remember also that 'sleep' spelled backwards is 'peels,' which is what you do to an orange, grapefruit and to a pear.

JAD: [laughs] I'm Jad Abumrad.

ROBERT: I'm Robert Krulwich.

JAD: And we'll see you later.

[ANSWERING MACHINE: First message.]

[HANNAH PALIN: Radiolab is produced by Jad Abumrad. Ellen Horne, senior producer. Lulu Miller, assistant producer. Production Executive, Dean Capello.]

[ANSWERING MACHINE: Next message.]

[HANNAH PALIN: Additional reporting by producers Ed Hepperman and Kara Oehler. Production support by Sarah Pelegrini, Scott Goldberg, Alaska Keville, Sam Lavender, Avir Mitra, Ryan Scammel and Jacob Wineburg. Thanks to the musicians we interviewed: Brad Creswell, Jeannine Dufree, Rob Christiansen, George Preston and Karen Havelick. And special thanks to me, Hannah Palin. You're welcome.]

[ANSWERING MACHINE: Next message.]

[HANNAH PALIN: Radiolab is supported by a grant from the Alfred P. Sloan Foundation. Radiolab is produced by WNYC, New York Public Radio, and distributed by NPR, National Public Radio. Support for NPR comes from NPR stations and the Geraldine R. Dodge Foundation, supporting educational, cultural and environmental initiatives to make our world more livable. On the web at grdodge.org. The John D and Catherine T. MacArthur Foundation, exploring how technology is changing kids and learning, at DigitalLearning.macfound.org. And the Skoll Foundation, connecting social entrepreneurs around the world with an online community called Social Edge. Learn more at SocialEdge.org. This is NPR, National Public Radio. Night, night.]

[ANSWERING MACHINE: End of message.]

 

-30-

 

Copyright © 2024 New York Public Radio. All rights reserved. Visit our website terms of use at www.wnyc.org for further information.

 

New York Public Radio transcripts are created on a rush deadline, often by contractors. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of programming is the audio record.