Play Live Radio
Next Up:
0:00 0:00
Available On Air Stations

Science of Slumber: How Sleep Affects Your Memory


This is SCIENCE FRIDAY. I'm Joe Palca, sitting in for Ira Flatow. If you add it up, we spend a lot of time sleeping, about a third of our lives, actually, and it turns out our bodies don't just power down as we slumber. Research is showing that sleep plays an important role in how our brains process and store the information that we learn throughout the day.

A study published in last week's edition of the journal Nature Neuroscience looked at how sleep patterns change as we age and how those changes affect our memories. The researchers found that as we got older, as we get older, the quality of our sleep deteriorates, and so does our ability to remember newly learned information.

So how much rest does your brain really need, and what does it do with that rest once it's getting it? We're talking about the science of sleep this hour. We want to hear from you. Maybe you have some theories about what the brain is doing while you're asleep. We'd like to hear about that. Give us a call. It's - the number is 1-800-989-8255, that's 1-800-989-TALK. And if you're on Twitter, you can tweet us your questions by writing the @ sign followed by scifri. If you want more information about what we will be talking about this hour, go to our website,, where you'll find links to the topic.

And now let me introduce my guests. First we have Matthew Walker, he's an associate professor and principal investigator of the Sleep and Neuroimaging Laboratory in the Department of Psychology at University of California, Berkeley. He's co-author of the study published last week in Nature Neuroscience, I guess it was actually a review article. He joins me from a studio in Berkeley, California. Welcome to the program.

MATTHEW WALKER: Thank you very much for having me.

PALCA: Also we have Robert Stickgold is associate professor of psychiatry and director of the Center for Sleep and Cognition at Harvard Medical School and the Beth Israel Deaconess Medical Center in Boston, Massachusetts. He joins me by phone today. Welcome.


PALCA: And you're still able to see - you're still aboveground there, the snow hasn't come to the roof of your house?

STICKGOLD: I can see all the way out to the street.

PALCA: OK, well that's good. Well have an update at the end of the hour. Next is Ken Paller, he's a professor of brain, behavior and cognition in the Department of Psychology at Northwestern University. He joins me from a studio in Evanston, Illinois. Welcome to the program.

KEN PALLER: Thank you, Joe.

PALCA: And we also have Michael Silber, he's a professor neurology and co-director of The Center for Sleep Medicine at the Mayo Clinic in Rochester, Minnesota. He's also pas president of the American Academy of Sleep Medicine. Welcome to you.

MICHAEL SILBER: Thank you, pleasure to be here.

PALCA: So we have a lot of people and a lot of things to talk about, but let's start with you, Matthew Walker. Tell us a little bit about what this review article was looking at. What is happening with sleep - I mean memory and aging and its relationship to sleep? I guess that's kind of a big question, but you can give us a short answer.

WALKER: It's a fine question, we've got an hour. There's actually two articles. One was a review article that myself and Robert Stickgold wrote, and the other was a scientific publication regarding aging. So regarding the aging, it's interesting, I think all of us know as we get older that our memory starts to deteriorate and is not as precise as it used to be.

But perhaps what people don't also understand is that one of the quintessential, physiological hallmarks of getting older is that our sleep starts to deteriorate. And that deterioration actually starts to happen quite early on. We can see it in the electrical brainwaves.

And what we found in this new paper was that those two things aren't actually independent. It's not that we simply get old, and memory starts to go, and sleep starts to deteriorate. But those two things actually are significantly interrelated. And perhaps one of the contributing factors to our poor memory as we get older is the deterioration of sleep, knowing now what we know about the importance of sleep for learning and memory.

PALCA: So maybe I can turn to you, Robert Stickgold. You've done a lot of research on sleep and memory. How does that work, that you can actually tell which stage of sleep is most critical for consolidation of memories?

STICKGOLD: Well, it turns out that probably all the stages of sleep are involved, but they're involved in different ways. And so what we will classically do is we'll train subjects on some memory task, and it might be a list of words, or it might be a typing sequence. So it can be very different types of memory problems.

And we'll train some subjects in the morning and test them 12 hours later that evening, and other subjects will train in the evening and test them after a night of sleep in the morning. And we'll compare how those two groups do.

And what we see pretty consistently is that the ones who got a chance to sleep will actually be performing much better after that 12 hours than the ones who had been awake. And then what we do is we go, and we look, and we say OK, well, what about their sleep seems to predict the improvement?

So on one task it might be the amount of deep sleep you get early in the night, and this would be the case more for things like verbal memory, that you'll see that the amount of improvement subjects show after sleep will depend on how much of that slow wave, that deep sleep they get, whereas in other tasks it might correlate with the amount of REM sleep that they get.

PALCA: I see, so it's time-dependent and stage-dependent, depending - and I should say that people maybe don't know that sleep has many stages. There's REM sleep, where your eyes are flopping around and other things happen, and then there's what they call non-REM sleep, which they break down into several categories of slow-wave and fast-wave. So...

STICKGOLD: Yes, so in a given night, you go through a 90-minute cycle. You go into very deep sleep, and then your sleep lightens up, and you come into REM or rapid-eye-movement sleep, which is notorious for its intense dreaming. And then you, in the next 90 minutes, you go back down to deeper sleep and back up to REM. And you do that all night long. You get about five of those cycles in a night.

PALCA: I have to say that we are having this discussion on the 60th anniversary, not to the day, but the 60th anniversary of the paper by Eugene Aserinsky Nathaniel Kleitman describing REM sleep in Science magazine. I printed it out just so I'd have something iconic here while I was doing the show.


PALCA: So Ken Paller, given that, is there anything in your research that shows that there's something about improving sleep that would help us with memory during the night?

PALLER: Sure, one of the dimensions that Bob almost mentioned is that there are different types of memory. So we want to be clear that these different types of memory depend on different brain mechanisms, and then in turn different aspects of sleep may be important for the different types of memory.

As Bob mentioned, slow-wave sleep is one of the most interesting types of sleep to connect to a class of memory we call declarative memory, which is recalling and recognizing facts and events that have happened to you. And so by improving slow-wave sleep, it might be possible to improve that type of memory.

PALCA: So what do you think? Is there going to be a pill someday that you can take before you go to bed or in the morning when you wake up that's going to improve not just your ability to sleep but the kind of sleep you get?

PALLER: I don't know if there'll be pills that'll have the right specificity. Right now the methods that are being used are to try to stimulate the brain during sleep, to try to see exactly what the connection is between these different stages and the memory effects that you can measure both by looking at the physiology of the brain at the time and also measuring memory after people wake up.

WALKER: And something to remember is that a type of sleep that might be good for one type of memory might not be good for another. So as you try to tweak your sleep one way or the other, you might be, you might be doing great - you might do better at remembering details of an event, but you might end up being poorer at abstracting the gist or the rules associated with it.

Those might be more useful interventions, though, in clinical populations who have deficits. So for example there's a type of neurophysiological event that we see in the EEG, what's called a sleep spindle, and patients with schizophrenia have about a 50-percent reduction in the number of those sleep spindles that they have.

So it might be that medications that can increase that would be of general value by getting them back into a normal range.

STICKGOLD: And just, too - sorry, I was just going to mention, just to deepen the plot, we also of course understand that sleep is not just only serving one particular function such as learning and memory. It serves a whole constellation of biological functions, both for the brain and also for the body. And so I think to think about how to exclusively brute-force perhaps one type of sleep because we think it's important for one type of memory, we probably do the evolutionary process a disservice thinking that it doesn't already know how to homeostatically tweak exactly what type of sleep it needs each and every night.

PALCA: Homeostatically tweaking is something I always like to take some time out to do each day.


STICKGOLD: Yeah, I make a point myself, religiously.

PALCA: Exactly. Michael Silber, I haven't forgotten about you because - but I wanted to hold you because I think you're going to be very valuable in helping to answer some of our callers' questions. And so I'm going to go take a call now from William(ph) in Hayley, Idaho. William, welcome to SCIENCE FRIDAY. What's your question?

WILLIAM: Thanks for having me on the show. I've been diagnosed as ADHD, and I guess I've been pretty bad ADHD my whole life. But I've always - most of my whole life, even my mother reminds, even when I was a child, I, you know, four hours of sleep is about what I get at night. And it's - I wake up fully rested, and it just sleep's not a real important issue for me.

PALCA: OK, so William, let me ask Michael Silber. If William is only sleeping four hours a night, is that a problem? Would a sleep disorders clinic want to have a look at him?

SILBER: Well, there is a wide range of the duration of sleep that people seem to need. Most of us need between seven and eight hours, but there are some people who seem to need less, some people who seem to know more. My first slight concern might be whether the medications that the caller is using for ADHD, which are really stimulant medications, could conceivably reducing his ability to sleep at night.

That doesn't seem to be so because he says he's had it since he was a child, and he's not very concerned by it. In general, if short sleep is not causing any major symptoms, we're a little less concerned. The person isn't tired during the day, isn't distressed by it. But just in general terms, and I'm not referring to the specific caller's issue at the moment, there is increasing and very interesting work that short sleep, less than five hours a night, epidemiologically in a large population is associated with medical problems, including glucose intolerance, which can give rise to diabetes, high blood pressure, and on huge epidemiologic studies, on an average, a shorter life expectancy.

Now, this doesn't apply to an individual person. I want to reassure the caller that he mustn't apply this data to himself. But this is very interesting that short sleeps seems, in general, not to be good for one in a metabolic sense, apart from the fact that in general people who are voluntarily depriving themselves of sleep, and that's not what the caller's doing, who are trying to sleep less because their lives are so bus and so full, well, also have memory problems and a whole lot of cognitive impairments the next day. And we recognize that very, very well in drivers and in others who have high-stakes occupations and don't sleep enough at night.

So it's a very important question.

PALCA: OK, well, we'll take it up again after we take this short break. We're talking about sleep and memory, and so stay with us. We'll be more - we'll be back.




PALCA: This is SCIENCE FRIDAY, and I'm Joe Palca. A programming note: Next week an asteroid will be passing very close to Earth, really close, I mean 17,000 miles or so, but that's close. Join Ira Flatow next week for the coverage, the live coverage of Asteroid 2012DA14 as it makes its closest approach. That's going to be live next week on SCIENCE FRIDAY.

This hour we're talking about sleep, and my guests are Matthew Walker, he's an associate professor in the Department of Psychology at University of California, Berkeley; Robert Stickgold, an associate professor of psychiatry at the Harvard Medical School; Ken Paller, professor of brain, behavior and cognition in the Department of Psychology at Northwestern University; and Michael Silber, professor of neurology and co-director of The Sleep Center - sorry, the co-director of the Center for Sleep Medicine at the Mayo Clinic in Rochester.

And I'd like to introduce one other guest who's on the line, David Dinges is director of the Unit for Experimental Psychiatry and the chief of the Division of Sleep and Chronobiology at the Perelman School of Medicine at the University of Pennsylvania. And he joins me by phone from Philadelphia. Welcome to the program.

DAVID DINGES: Good afternoon.

PALCA: So maybe I could just start by asking you, our last - just before we left for a break, we were talking with a caller who was talking about getting only four hours of sleep a night and contrasting that - he says everything's fine, he doesn't have any problems - contrasting that with people who are highly sleep-deprived.

You have worked with this problem of sleep deprivation. Tell us a little bit about how big a problem that is, both physiologically and psychologically and sleep-related - with relation to sleep.

DINGES: Well, sleep occurs in the time domain, as we all know, and many people moan about the fact that they have to spend too much time sleeping, and they'd rather be doing other things, and as a result we live in a culture where commute time and work hours and social time steal our sleep. And the question is how far can we go with that an not pay a price.

And the answer is when we start to reduce our sleep down to shorter levels, below six hours, six or less, we start to see increasing signs of dysfunction in the brain and physiologic signs in the body, regulation of food and that. So we have to be careful about reducing it.

PALLER: So we have to be careful about reducing it. The caller may be one of those naturally very short sleepers. They are quite rare, but they do occur. But it's also true that they may not be aware, and this is the big problem, they may not be aware of the deficits they're experiencing with the sleep loss. That is to say they may feel OK and be motivated, but in fact they may be at risk for falling asleep behind the wheel or not remembering things as well or in general just not functioning at their peak level because they're not getting enough sleep.

And there's a lot of people, as I understand it, who say oh, I only sleep four hours a night, and yet when they come into a lab where they can actually be carefully measured or wear a device that can whether they're moving or not, in fact they do sleep longer than that, they're just not conscious of it.

DINGES: Yes, they'll say that they can do it, and we've studied literally about 600 people through four hours a night sleep repeatedly, and most of them develop - about 75 percent will develop a significant deficit in functioning over time, even though again their full awareness of how bad it is is not very accurate as it gets chronic day after day.

But I will also tell you that there are a few people who can tolerate it. Even though they sleep seven or eight hours a night, or six and a half, they can tolerate the sleep loss better than others. So one of the mysteries right now is not only how much sleep do you need but what is your tolerance or vulnerability to the sleep loss.

Everybody will get impaired, but some people get impaired much faster than others.

PALCA: Go ahead.

WALKER: Yeah, I was just going to say...

PALCA: This is Ken Paller, right? No, no, I'm sorry this is Matthew Walker, sorry.

WALKER: That's OK.

PALCA: It's a lot of people to keep track.

WALKER: I would like to be Ken Paller, but that's - I think it's interesting and perhaps not surprising that people do suffer, the majority of people do suffer so easily when you take sleep away from them because in some ways it's never a challenge that evolution has had to face in terms of solving and placing safety nets there for because human beings are one of the few species that will deliberately deprive themselves of sleep.

So unlike the challenge, for example, for energy, our bodies actually have interesting mechanisms for storing energy in adipose tissue with fat to get us through those times when perhaps food wouldn't be abundant. Now that's not a problem we face anymore in modern-day society, but the sleep, sleep is an interesting issue in that perspective that perhaps evolution hasn't needed to face that challenge because it's not something that's common across a broad array of species.

And we human beings, we decide to engage in that type of behavior, and there doesn't seem to be good biological mechanisms immediately obvious to us that can hold onto the challenges of sleep deprivation, and hence we fall very quickly.

PALCA: OK, let's see if our callers can join the conversation, and let's go first to David(ph) in, well, North America. David, welcome to SCIENCE FRIDAY.

DAVID: Thank you, Sacramento, actually, and I must say right off the top go Bears.

PALCA: OK, yeah, well, that's nearby, yeah.

DAVID: I've often wondered how - kind of how late in the day people can really be able to sort of navigate a night of sleep and still have a coffee, you know, in the afternoon or the evening. I know people that actually, they'll have a caffeinated drink at, you know, 7, 8, 9 o'clock at night and sleep well. And what's the mechanism in our body that can handle the caffeine? For me...

DINGES: There's a biological basis for differences in responding to caffeine. Some people carry a genetic variation, just a natural genetic variation in the genes that caffeine binds to that permit them to clear it from their system rapidly, while others it takes longer to clear. And as a result, there are these differencs in how well people tolerate caffeine before they sleep at night.

But the other thing I wanted to add was aside from total sleep time, to go back to sort of timing, the key thing about sleep is timing. It's a circadian behavior. It's intended to occur at night. Nightshift work, jet lag is incredibly challenging physiologically because sleep itself is displaced and disturbed from that. So it's duration and timing and quality.

PALCA: OK, thanks for that call. Let's go now to Ken(ph), Ken in Dyersburg, Tennessee. Welcome to SCIENCE FRIDAY, you're on the air.

KEN: Does the practice of meditation affect memory and sleep?

PALCA: OK, meditation and sleep? Robert Stickgold, maybe that's something you can talk about?

STICKGOLD: I can talk about but I can't answer. We've actually tried to do this study twice. The question we were trying to ask is whether meditation can replace sleep in terms of providing a brain state that allows for that kind of offline memory processing, and for purely technical reasons, both of those experiments never quite got as far along as we had wanted.

So we don't really know. The really top-notch Nepalese-type meditators who do this as a lifestyle report that they need very little sleep. But I don't know that there have been any good studies to confirm that. So whether it can do the memory processing remains an open question, and whether it reduces the total amount of sleep you need unfortunately is still an open question, unless somebody else knows something more recent than me.

WALKER: I think the only other thing...

SILBER: Well, from a different viewpoint, patients with insomnia who have difficulty sleeping, relaxation techniques of various types have been shown in studies to help induce and support sleep so looking not specifically at meditation but relaxation, the whole range of relaxation techniques are very helpful for people with insomnia, starting from simply unwinding before sleep and then going through some very, very formal processes.

And I think in that way people can be helped when they have insomnia, but that's a little different from the question of does it affect or replace normal sleep.

PALLER: I mean this is a Holy Grail issue: Is there an activity that involves consciousness that does not involve sleep that would replace it? And in my experience, we like, Bob Stickgold and other labs, have searched for it, but we've never found anything that comes close to it. Sleep is biologically demanded by the brain and body and programmed to occur.

SILBER: I should also say, Matt Walker, that people have actually studied the patterns of brainwave activity during meditation and also placed people inside a brain scanner and actually had them meditate, surprisingly, in a brain scanner. And what you see is that the biology, as it were, of meditation, the brainwave activity patterns and the structures that turn on and off, are not the same as the stage of sleep that we at leepsauctures that turn an

So that's not to suggest that meditation couldn't have an additional benefit on memory and it couldn't perhaps affect sleep. All those things are possible. But if it did, my suspicion is that, for example, in terms of memory, it would be achieving those memory benefits in different ways based on the unique biology that is meditation versus sleep.

PALCA: So this is something - I'm glad we've kind of gotten to this because I want to know, then, if sleep is critical or at least very important for memory consolidation or long-term memory storage or call it what you will - is that the function of sleep? In other words, is that why we sleep?

STICKGOLD: You know, that's a funny question because I think we spend too much time asking. I often give the example of two tongue researchers, where one insists that the function of the tongue is to taste food, and the other insists that the function of sleep is to articulate - of the tongue is to articulate speech. And obviously the tongue is doing both of those things, and sleep is doing several things as well.

It's involved in immune functions, it's involved in endocrine regulation and it's involved in memory. And I don't know that it's right to talk about the function, although I would tend to argue that memory is the only one we've come up with that has a strong argument of why you have to become so disconnected from the rest of the world.

It turns out that the human brain is not like a VCR, you know? In a VCR you can be having the VCR record one channel while it's showing on the TV a second channel. We can't do that. If we're trying to process our memories, we can't have information coming in at the same time, or they just conflict with each other. People will do that even now as they're listening to your show.

Someone will say something and they'll start thinking about that, and to do that they have to cut off the input of the other people talking. So it's probably the memory function that requires the shutting off of conscious awareness of the outside world. But again, that's only a small part of what's happening while we're asleep.

PALCA: So David Dinges, I know you have to leave, but I want you to take a crack at that question too. I mean, is there - if you have to say to somebody why we sleep, I guess somebody said, then why are we awake? But, you know, that's a epistemological question, I think, or metaphysical. But do you have an answer for why we sleep?

DINGES: Well, I don't have the final formal, absolute answer, but I can say the problem's with the word we in that sentence. Humans like to know why they sleep as a species. The answer in this lies, why do all the animals sleep? Why is sleep so pervasive in the animal kingdom, but why are some nocturnal and diurnal? Why do some sleep in polycyclic patterns and some monocyclic?

It's highly likely sleep evolved out of the need to adapt to an environment changing in the circadian way. And as a result, it's taken on many functions, not one or two or three. Because we're a very symbolic creature that thinks symbolically and relates to the world by structuring symbolic representations of the way the world is and causal relationships, we value memory enormously.

But to the tiger, sleep exists so the tiger is faster than the antelope and can sneak up and hide and attack it and eat it. In other words, sleep serves functions for all animals: physiological, developmental functions, learning functions, attention functions. I think it's us who are obsessed with it, has to have one function, it has to be one thing, but in fact it seems to, as nature often does, ensure that it does many things.

As Dr. Walker was suggesting, it handles metabolic control. It probably protects our ability to deal with stress. It copes with immune response that's affected(ph), and a host of things. So we're probably getting enormous benefits from adequate sleep every day, and this constant war on sleep - to try to eliminate it or shorten it or find a substitute for it - is, I think, misguided.

PALCA: Well, David Dinges, I know you want to go, so I'll just say it was nice to talk to you again.

DINGES: Thank you very much.

PALCA: And David Dinges is director of the Unit for Experimental Psychiatry and chief of the Division of Sleep and Chronobiology at the Perelman School of Medicine at the University of Pennsylvania. I'm Joe Palca, and this is SCIENCE FRIDAY from NPR.

We're talking about sleep and memory and taking your calls at 1-800-989-8255. And let's take a call now from Steven(ph) in Beavercreek, Ohio. Welcome to SCIENCE FRIDAY. You're on the air.

STEVEN: Thank you. My question is about taking naps during the day.


STEVEN: Does it have an adverse impact on the quality of sleep during the evening, especially in an older population? Thank you.

PALCA: OK. Interesting question. I'm a little unsure who to put it to. Any volunteers?

SILBER: Shall I have a go at it?

PALCA: Sure. This is Michael Silber.

SILBER: Napping is very variable, and it depends on a lot of factors. If one's sleep-deprived at night and one's really not getting enough sleep at night, taking a nap during the day may be very restorative, may not be the best way of doing it, but it is one way of doing it.

Now, we're used to the concept that our sleep should be highly consolidated at night. Our society puts a lot of store on that, that you go to bed at a regular time, sleep through. If we wake up once or twice, we feel we've got insomnia. We get up in the morning. We shouldn't nap during the day.

But not all cultures follow that. The cultures that have followed the siesta, for instance, people stay up much, much later at night, and then it's understood there will be a long nap in the early afternoon. So there are cultural changes.

There's very interesting work from, historically, from the Middle Ages, which suggests that people went to sleep close to sunset, then woke up in the middle of the night and were up and about for a couple of hours sometimes doing - interacting with neighbors, et cetera, then going back into what was called second sleep. And that was considered entirely normal at that time.

So we've got to understand that what we think is normal is in our society normal, but not the only way that the human brain and humans can deal with sleep. If you're having trouble sleeping at night, taking a nap is not generally a good thing because it can be subtracted out of the sleep at night. Generally, people have to experiment themselves.

If they're having insomnia, napping generally we try and avoid as much as possible, but it's really an individual thing, taking into account any sleep disorders one has, the amount of sleep one has at night, one's normal circadian tendency.

STICKGOLD: This is Bob Stickgold.

PALCA: Yeah. Go ahead, Bob.

STICKGOLD: I just want to add that we've done some studies looking at naps in terms of the memory processing and have been rather stunned, really, by the fact that in almost every experiment that we've tried, an hour-and-a-half nap seems to do as much good for memory processing as an entire night of sleep, and we continue to ponder that and sort of conclude that OK, we just don't get it yet. But in studies where six hours of sleep at night seems not enough to lead to consolidation of memory of a particular task, an-hour-and-a-half nap will. So there's something, at least from the memory perspective, rather magical and unusually efficient about napping as opposed to nocturnal sleep.

PALCA: Dr. Stickgold, I also think your battery might be dying on your cordless phone, if that's what you're talking to us on. So - or move closer to your bay station because...

STICKGOLD: OK, I'll get closer.

PALCA:'re breaking up a little bit. We have to take a short break but when we come back, we're going to talk some more about sleep, memory, learning. And I still want to challenge our - my guest to talk about REM sleep and all this, because that's another strange thing where we certainly at least seem to have a lot of mental activity going on, and I wonder if that is helpful for remembering things or not. Certainly dreams have been an important part in our psyche and our culture. Maybe they're important in our learning and memory. Anyway, we'll be back and talk about that after this short break. This is SCIENCE FRIDAY from NPR.


PALCA: This is SCIENCE FRIDAY. I'm Joe Palca. We're talking this hour about how sleep affects how our ability to store and process memories. My guests are Matthew Walker, he's and associate professor in the department of psychology at the University of California, Berkeley; Robert Stickgold, he's at Harvard Medical School and Beth Israel Deaconess Medical Center in Boston; Ken Paller, professor of brain, behavior and cognition in the department of psychology at Northwestern University; and Michael Silber, professor of neurology and co-director of the Center for Sleep Medicine at the Mayo Clinic in Rochester. And we're interested in hearing your calls about sleep and memory, 1-800-989-8255, 1-800-989-TALK.

And I threw out this question of REM sleep and memory. Is there - I mean, earlier we were talking about different phases of sleep having different importances at different times in the night. Does anyone want to tackle the question - maybe you, Matthew Walker, about what REM is doing there in terms of this memory process?

WALKER: Yes. I think as we spoke about before, what we'll find is that REM is multifunctional in terms of the different processes that it serves. One of the things that's been emerging over the past five or six years in the literature is that REM sleep seems to serve an almost sort of overnight therapy benefit in terms of our emotional well-being and our mental health. And what's interesting is that REM sleep chemically, in terms of the anuric transmitters that swirl around the brain, REM sleep is perhaps the only time during the 24-hour period where a particular stress neurochemical called norepinephrine or noradrenaline is actually suppressed and it's completely shut down from the brain. And it seems to be that REM sleep is a perfect chemical environment, almost like a therapy session, where you can take emotional events from the prior day and perhaps just smooth the sharp edges off those experiences, and almost like an overnight soothing balm, as it were.

And there's now scientific evidence to suggest that in fact REM sleep does provide that type of dissipation, that detoxing of emotional experiences, and it's perhaps no surprise that we all have that feeling of, you know, if troubled, get to bed, you'll feel better tomorrow. So I think that that's an emerging idea and it jibes very well with the clinical literature. For example, we know in PTSD, which is perhaps one of the quintessential psychiatric conditions of dysfunctional emotional memory processing, there you see profound disruptions of sleep, including REM sleep. And by trying to ameliorate(ph) or restore that REM sleep, recent findings have demonstrated benefits in terms of clinical PTSD outcome. So I think it's a fascinating area that we still don't fully understand enough about but certainly suggests that that may be one of the functions of REM sleep, this idea of, you know, the best bridge between despair and hope is perhaps a good night of REM sleep.

PALCA: Ken Paller, do you want to take a crack at that? is that your sense of REM as well?

PALLER: Yeah. Sure. And I can make a connection between that and the memory story. So the idea that Matt was talking about that we're doing some emotional regulation, perhaps, the current concerns of the day that the emotional things that are bothering us are something we work on through the night and there's evidence that the different REM periods are involving a progression of that is we think about the recent events and relate them to all the memories that are related to the same idea and might help us solve the problem. So these higher principles of thinking about current problems might be what are recruiting both the processing during REM and perhaps some of the memory processing that happens in other stages of sleep as well.

PALCA: Let's take another call now - Oh, sorry, go ahead. who's that?

STICKGOLD: Bob Stickgold.

PALCA: Oh, hi, Bob. Yeah. Go ahead.

STICKGOLD: I was just going to comment that there's also evidence that suggest that deeper sleep, that what we call slow-wave sleep, non-REM sleep, might be more involved in sort of strengthening memories in a way that they're for REM. So for example, if you want to remember someone's name or phone number that might be where that non-REM sleep is valuable. And that REM sleep seems to be more involved in memory processing where you're not trying to exactly remember it in the form that you first saw it but rather to extract meaning from it. So to get patterns or to figure out the rules of something or to extract just to get the sort of the executive summary. But that seems to be more of what REM sleep that's about and that would fit in with the emotional piece too because, of course, if something had happens during the day, you don't simply want to remember it better than next morning. You want to understand it.

You want to relate it to other things in your life and see how it fits into your associative networks and, sort of, process it that way. And that seems to also be part of what REM is doing.

PALCA: OK. Bob Stickgold, thank you for that. Let's go to a caller now and take a call from Collin in Ridgeville, Indiana. Collin, welcome to SCIENCE FRIDAY. You're on the program.

COLLIN: Thank you. My question is regarding a - not the lack of sleep, but too much sleep. The lack of sleep has been well-studied and researched, it seems like. However, I'm very well aware of the physical effects of commonly known Sleeping Beauty syndrome, but I'm wondering what the psychological effects can be and if that's even been studied. Unfortunately, I have not been able to find any material on that.

PALCA: Anyone? Matt Walker? Matthew Walker?

WALKER: Maybe (unintelligible)

PALCA: Hypersomnia or...

WALKER: Yes. So hypersomnia has been something that's been discussed in the literature, in depression, for example, this idea of sleeping too long. What's interesting is that the recent evidence that has tried to, sort of, place a scalpel and separate out, is it really sleeping too long, or is it staying in bed too long?

And when you do those types of more careful approaches and analysis, I think there remains some controversy, at least in the depression field, as to whether it's really simply too much sleep, hence hypersomnia; or in fact, it's just too much time in bed yet with a perhaps normative amount of sleep. So that's in the depression literature, but perhaps I'll leave it to some of the other clinical folks to discuss the more general issue of hypersomnia.

PALCA: Michael Silber, do you see a lot of hypersomnia patients in your clinic?

SILBER: Well, we see lots of hypersomnia in the sense of people who cannot keep awake during the day. And usually, there's a clear answer to that: either insufficient sleep at night, medications, or drugs or diseases such as obstructive sleep apnea or narcolepsy. Long sleepers are very interesting. You remember, I was talking a little earlier about the ill effects of too little sleep.

Actually, multiple studies have shown, epidemiologically, that it's a u-shaped curve, and that sleep that is too long, beyond nine or 10 hours, also seems to have - be related to reduced mortality. And whereas we understand a great deal about the short sleep, we understand very little about the effect of too much sleep.

PALCA: So I'm sorry. Did you just say that too much sleep reduces your mortality?

SILBER: Increases mortality, I'm sorry. It's the u-shaped...

PALCA: Oh, OK. I just want to make sure I had the right - you going in the right direction.

SILBER: Yes, thank you for correcting that. But too little and too much seems to be associated with increased mortality. Now whether that is simply that people with too much sleep have a whole lot of chronic illnesses, fatiguing illnesses - and it's the illness that does it rather than the sleep, is completely uncertain at present, a lot less certainty about this. But there are some people who just have long sleep, just as there are people who have short sleep. And if they get 10 hours sleep at night or 11 hours, they feel fine the next day.

If they get less, they feel sleepy. And we call them long sleepers, but we don't understand them very well at all. Presumably, it's a genetic variability. But people who really sleep too much at night probably - and still feel tired during the day probably have a disorder. And there's a spectrum of disorders, including something called idiopathic hypersomnia, that are not well understood yet.

PALCA: Right. Collin, you mentioned Sleeping Beauty, and I guess the answer here has more to do with Goldilocks, where you don't want too little - you don't want your porridge too hot or too cold. You want it just right. So that seems to be the answer there. Thanks very much for that call. Let's take another call now and go to Adam in Fresno, California. Adam, welcome to SCIENCE FRIDAY. You're on the air.

ADAM: Great. Thank you for taking my call. I had a question about polyphasic sleep. And...

PALCA: OK. Now you have to define that. Somebody else mentioned that, and I'm remiss for not explaining what polyphasic sleep is. How would you define that?

ADAM: I guess, from my limited understanding, is structured napping.

PALCA: OK. Is that - hold on just a second. Now, can we all agree that that's an appropriate use of the term?

STICKGOLD: Polyphasic just means that you sleep more than one time during 24 hours.

PALCA: OK. Bob Stickgold, thank you for - so go ahead, Adam. I didn't mean to interrupt, but I just wanted to make sure we were all talking about the same thing.

ADAM: Sure. And if there is any research on increasing productivity during the day or of it actually decreases cognitive ability by either - some of the examples I saw were four 30-minutes naps every six hours or six 20-minute naps every four hours. And if you just have any experience or...

PALCA: Oops, sorry. Did I cut you off there?

ADAM: Oh, no.

PALCA: I'm sorry I did that. Go ahead. You were saying, if there are any evidence for what, I'm sorry?

ADAM: If there's any evidence for an increase in productivity by switching to kind of a more polyphasic sleep schedule.

PALCA: OK. Any takers on that one?

STICKGOLD: So there's this new fad, if I may call it that, that I've noticed amongst college students, which is this concept of trying to spread your sleep out over many, many short - many more, but shorter bouts of sleep in the day. And I don't think there's any evidence that that's particularly helpful. When I hear students talk about it, they're trying to actually come up with a method of reducing their total sleep time. And I think David Dinges, who is gone now, has actually done a study showing that you can't - if you take a nap during the day, it doesn't mean you need less total sleep in 24 hours.

There are some benefits from naps, just in terms of sort of, I don't know, I often describe it as emptying your inbox and sort of getting you back up if you're feeling burnt out. But whether that's because you weren't getting enough sleep the night before or some real special benefit of napping, I don't think that studies have been done that can answer that.

WALKER: I think also - sorry. Matt Walker...

PALCA: Go ahead. Sure, man.

WALKER: speak about that. If you look at how humans tend to want to sleep, it seems to be either, you know, sort of a monophasic way or at least a biphasic way, where there's, perhaps, a long bout during the night and then maybe a siesta-like pattern during the day.

But certainly, what we don't adopt is as adults is this highly polyphasic sleep cycle and naturally is what I mean. And I think if it had such extraordinary benefits to be sleeping at a polyphasic way, evolution would have naturally sailed off, sort of directional development along that pathway. The fact that we don't have that biologic pressure to have highly polyphasic sleep, I think, probably tells us something in terms of, truly, whether it's useful or not.

PALCA: We're talking about sleep, sleeping and memory, and how those two things are related. I'm Joe Palca and this is TALK OF THE NATION: SCIENCE FRIDAY. And let's take one more call now and go to Bryan(ph), Bryan in Brighton, Michigan, I guess. Welcome to SCIENCE FRIDAY. You're on the program.

BRYAN: Thank you for taking my call.

PALCA: Sure.

BRYAN: I have a lot of questions and I don't want to get too in-depth with it. But I had insomnia, and then later on, developed PTSD from being overseas. And I've always had a problem with sleep. I've done sleep studies. On overage, I will get, literally, one to two hours of sleep a night until about the fourth or fifth. And then, I'll start getting double vision, tremors, slight-like sound hallucinations. And I've done everything under the sun for medication-wise. And also, I've gotten, you know, a lot of friends - I was in sleeping there. They've taken the TV out of my bedroom. And the only thing I associate with my bed is sleep. I guess, my question is, is there any thing out there that, you know, I might be missing that could help me get sleep. I have trouble, actually, going to sleep. I just can't get to sleep.

PALCA: OK. Bryan, you understand that we can't give you any specific medical advice, but maybe Michael Silber can talk about what's in the pipeline or what new things are becoming available.

SILBER: Sure. Insomnia is a symptom, not a disorder. So when first - when dealing in general terms with a situation of severe insomnia, one wants to ask, what are the causes, what are the contributing factors. You mentioned that, unfortunately, you have had experiences which resulted in PTSD, which is, certainly, going to be a contributing factor, but also that your sleep was bad for as long as you can remember. So one would want to, in general terms, look at anything that disturbs sleep: how much depression they might be, anxiety in a person, any physical factors affecting sleep, such as the condition of restless legs, pain. There are many things inside the body, outside the body that can affect sleep.

So the first approach we would generally take is to try to work out what are the contributing factors to this problem of insomnia? Then, once - if we've corrected those factors or none of them are particularly correctible, we would then have generally two approaches. The one you've tried clearly, which is sleeping tablets, hypnotics, many of them work, but they don't work perfectly. And sometimes, the expectations may be a little more than we can achieve with them. There are new sleeping tablets being developed. We're going to have new mechanisms of sleeping tablets available. And with time, hopefully, we'll get sleeping tablets that have better effect with fewer side effects.

The other and generally more helpful approach you've alluded to, and that's what we call cognitive behavioral therapy, which is a series of well-proven techniques, well-studied, shown to work in controlled trial to correct problems of insomnia by behavioral approaches. And you've mentioned some of the things that you've learned to do. Now, obviously, I don't know to what extend you've really - I've gone through a formal program with a behavioral psychologist who is trained in sleep psychology. But if you haven't had a really formal program, that might be another approach that people with severe insomnia can do. Self-help works up to a point, but sometimes, one needs the help of - ongoing help with a specialized psychologist. We don't have a perfect solution to insomnia. And I'm really feel so bad about people with severe, intractable insomnia. But generally, with combinations of things, we can provide some help.

PALCA: OK. Well, thanks very much for that. Ken Paller, I'm going to give you the last word. And I - we don't have a lot of time, but I just wonder if you could quickly tell me, where do you see the field of sleep research going? Is this going to be - is sleep and memory going to be the focus? Or do you see something else coming along that's very important?

PALLER: Well, I'd like to take the view that's contrary to some of the views we've heard earlier, that maybe sleep isn't optimized already by whatever factors have led to where we are now. And that there are possibilities for, maybe, improving things, improving memory processing during sleep, for example, is one thing. And so I think together with the clinical methods Michael has mentioned, there are possibilities for making sleep better and in the course of doing that, improving the various benefits of sleep on arousal and, perhaps, on memory as well.

PALCA: All right. Well, I'm afraid we have to leave it there. It's been a fascinating hour. Thank you all for joining me. Matthew Walker is an associate professor in the department of psychology and principal investigator of the Sleep and Neuroimaging Lab at the University of California, Berkeley. Robert Stickgold is an associate professor of psychiatry and director of the Center for Sleep and Cognition at Harvard Medical School and at Beth Israel Deaconess Medical Center in Boston, Massachusetts. Ken Paller is a professor of brain behavior and cognition in the department of psychology at Northwestern University in Evanston, Illinois. Michael Silber is professor of neurology and co-director of The Center for Sleep Medicine at the Mayor Clinic in Rochester, Minnesota. So, gentlemen, all, thank you for joining us.

I know there are times when I wish I had that for my children.

STICKGOLD: Joe, get a good night sleep.


PALCA: OK. I'll do that. Thank you.

WALKER: You sleep well.

PALCA: I'll work on it, believe me.

In Washington, D.C., I'm Joe Palca. Transcript provided by NPR, Copyright NPR.