Dr. Matthew Walker on Sleep for Enhancing Learning, Creativity, Immunity, and Glymphatic System
[Rhonda]: Today’s guest is Dr. Matthew Walker,
author of New York Times best selling book
Why We Sleep: Unlocking the Power of Sleep
and Dreams, Professor of Neuroscience and
Psychology at University of California Berkeley,
and Director of the Center for Human Sleep
YOU WILL NOTE… This episode starts sort
of abruptly. The opening discussion is about
HOW SLEEP PATTERNS CHANGE DURING HUMAN DEVELOPMENT,
especially around age 12 months, when motor
skill learning increases.
Matt and I had planned on doing an intro but
the camera was rolling and the discussion
was so interesting that we just kept going.
This is a good one, folks. Enjoy!
[Matt]: What’s really interesting is if you
look at sort of sleep, and we’ve done some
of this work, and I maybe will speak about
it on sleep and motor skill learning, and
that seems to be more dependent on this sort
of lighter form of non-REM sleep Stage 2 and
particularly the burst of electrical activity
the sleep spindles, there’s a really bizarre
bump in Stage 2 and sleep spindles during
development, it happens right around the 12-month
period, which is where, all of a sudden, you
start to see considerable limb or multi-limb
coordination. In other words, you start to
perambulate, you start to learn how to walk.
It’s almost as though there’s like a homeostatic
response, which is that with the drive for
motor plasticity and learning comes a response
from sleep to say, “Oh, now we’re into motor
skill learning? We need to consolidate.” And
you get this…it’s a really strange bump,
and it dies away again after the…
[Rhonda]: What’s the sleep, the spindles,
sleep two spindles, what stage is that?
[Matt]: Well, you see sleep spindles throughout
almost all of non-REM. So once you get past
the lightest sort of Stage 1 on REM, then
you get spindles throughout all of non-REM,
but they are a prototypical feature of Stage
2 non-REM as well. Then Stage 3 and 4, which
is sort of like the deep sleep stuff, you
also get spindles there, too. But in our hands,
at least, the strongest sleep stage and the
strongest electrical signature in your EEG
that is predictive of your motor skill learning
is Stage 2 and sleep spindles, both of which
seem to have this bizarre sort of, you know,
coincidental spike right around this developmental
phase of crawling, standing, walking.
[Rhonda]: Yeah. So many things going on during
the development. I mean, it’s so fascinating
to observe for sure.
[Matt]: And language, too. I mean, when we’ve
looked at this as well with teaching adults
foreign languages, or even actually just mathematical
languages or artificial grammar, sleep is
a huge component in that. But they also saw
this fascinating thing with development, which
was about not just concrete learning of individual
facts but the generalization of knowledge.
So this is sort of the thing that I think
separates us from computers, at least for
now, which is that computers are very good
at storing individual bytes of information
in a vertical way very clearly. And we can
do that, too, storing individual facts. But
what computers don’t do, which is what we
do, is intelligently integrate and associate
them together so that we can extract overarching
patterns and schemas and statistical rules
about this thing called the world in which
And, yeah, sorry, I was just gonna say that
you can, with infants, you can teach them
just these novel sounds, and each one of these
strings of sounds is unique and different.
But there’s something common about the grammar
that is binding and overarching across all
of them. There is an overarching schema that
you could learn in addition to each one of
the individual facts.
After they’ve had a nap versus an infant that
has not had a nap, post-nap, the infants have
extracted and understood the generalized rules
of what they’ve been learning, not just the
individual facts. Whereas infants that have
learned but haven’t napped have not sort of
made the abstraction.
[Rhonda]: So my being on nap nazzi, I should
be rewarded for that, right?
[Rhonda]: It’s absolutely like…
[Matt]: And it stays from the, you know, infancy
all the way through to adulthood. So if you’re
napping, I’m not gonna…well, there’s a few
double-edged sword aspects of napping.
[Rhonda]: So I wanted to ask you a question,
you’re talking about basically being able
to connect the dots, you know, how would that
sort of differentiates us from what a computer
can do, I mean, one of the things. But is
there a certain…I remember reading in your
book, and this is probably the part of the
book that I was more skimming, you know, like…and
that was the importance of REM sleep and dreaming
and creativity. So is that, connecting…because
I feel like connecting the dots, you know,
requires some creative thought to be able
to kind of, like, put things together and
come up with a big-picture idea and figure
things out. So is it known? Is REM sleep important
[Matt]: Yeah, it does seem to be. So if there
is…So, I mean, we could take a step back
and think about how does sleep achieve memory
processing, learning, information processing.
And sleep seems to be important in at least
three ways. First, you need sleep before learning
to actually get your brain ready to initially
soak up new information, to initially lay
down new memory traces.
But you also then need to sleep after learning
to take those freshly-minted memories in the
brain, particularly in a region that we call
the hippocampus, which you could think of
almost like the informational inbox of your
brain, but it’s very good at receiving those
sort of new memory files. But you need sleep
after learning to take those new memories
and then essentially hit the save button on
them so that you don’t forget those informational
pieces of the puzzle. So sleep before learning
to get your brain ready, to acquire information.
Sleep after learning to hold on to those individual
[Rhonda]: So let me get this straight. So
if you want, for example, short-term memory,
right, because if you’re sort of wanting to
store things in the hippocampus, even short-term,
that would be sleep before.
[Matt]: That’s right.
[Rhonda]: And then if you want to then consolidate
it and have a long-term memory, that would
be your sleep after?
[Matt]: Sleep after.
[Matt]: So you can’t cheat sleep on either
side of the memory equation. You’ve got to…You
can’t pull the all-nighter and hope to be
able to continue to learn. And we did this
study. We said sort of, you know, “Is it wise
to pull the all-nighter before the exam?”
So we took a group of individuals, assign
them to one of two groups, a sleep group and
a sleep deprivation group. A sleep group,
they get a full eight hours of shut-eye that
we measure here at the sleep center. The deprivation
group we keep awake all night under full supervision.
And they don’t get any naps. There’s no caffeine.
It’s miserable for everyone involved.
[Rhonda]: Wow. No caffeine.
[Matt]: No caffeine at all. And then the next
day, we place them inside an MRI scanner,
and we had them try and learn, and cram, essentially,
a whole list of new facts into the brain,
into the hippocampus. And the first result
was that the sleep deprivation group was about
40% more deficient in their learning ability.
So they learned 40% less, four zero, which
[Rhonda]: That’s astronomical. That’s a huge…
[Matt]: Non-trivial. I mean, if you want to
put that in context, I guess it’s the difference
between acing an exam and failing at miserably
40%. What was interesting, though, is what
was going on in the hippocampus, this informational
inbox of the brain. When we looked at that
in those people who’d had a full night of
sleep, you saw lots of healthy learning-related
activity. It was beautiful.
In the sleep deprivation group, we actually
couldn’t find any significant signal whatsoever.
And so it was almost as though sleep deprivation
had sort of shut-down-your-memory inbox and
any new incoming files were being bounced.
And we put forward a theory as to why that
was. Perhaps that the hippocampus being a
short-term reservoir of memory has a limited
storage capacity, perhaps a little bit like
a USB stick.
And you have maybe, in humans, a 16-hour recording
capacity for information acquisition before
you have to sleep. Because it’s during sleep,
then after learning. So that’s sort of the
story before learning. It’s not great. We
can show it. We know in the brain what part
of the brain is failing to produce those impairments.
[Rhonda]: In that study, you were testing
the ability to acquire new information.
[Matt]: Exactly, to sort of lay down those
fresh memories and just to grab hold of them.
And you can’t do that well without sufficient
sleep. And that seems to be in part related
to your non-rapid-eye-movement sleep or your
But then, what we’ve also done in lots of
these, and we and lots of other people have
now replicated this finding, sleep after learning
then takes those memories and it sort of it
hits the save button on them. It’s a little
bit crass. Actually, what really happens is
that during sleep, there is a file transfer
mechanism that takes place at night, that
we shift memories from that short-term vulnerable
storage reservoir, the hippocampus, and we
move them out to the long-term storage site
within the brain, which is the cortex, which
essentially acts like a hard drive.
And that means that when you wake up the next
day, there are two delightful benefits. First,
having shifted those memories from the short-term
vulnerable reservoir to that more permanent
sort of safe storage haven in the brain. They’re
protected, and they’re safe, so that you’re
going to remember rather than forget.
The second benefit, however, is that sort
of having cleared off those files from the
hippocampus, almost like shifting files from
the USB stick, you’ve cleared out all of that
fresh memory encoding reservoir, so that when
you wake up the next morning, you can start
acquiring new files all over again. So it’s
this sort of elegant, symbiotic system of
memory that happens.
[Rhonda]: Yeah, beautiful.
[Matt]: Yeah, absolutely. Sorry.
[Rhonda]: We’ll, of course, use all this.
[Matt]: I’m ethically sealed[sp]. It’s bizarre.
[Rhonda]: I remember reading somewhere that
when you sleep, and this is related to what
you were just talking about, that your brain
sort of replays, like, electrical-activity-wise,
it looks like you’re literally, like, reliving
the same thing you just learned or something.
[Matt]: That’s absolutely correct. So these
studies firstly happened in animals, and we’ve
now been replicating some of them with brain
imaging, with MRI scans in humans. Hard to
believe, but we can do it. But the original
findings were fascinating. They would place
electrodes into that structure that we spoke
about, the hippocampus, which also rats have
as well, and it helps rats learn spatial navigation.
And they would place lots of electrodes into
this part of the brain. And as the rat would
run around the maze and learn the maze, individual
cells would fire, and they would spatially
code which part of the maze the rat was in.
So different cells are mapping different parts.
It’s like sort of navigating a route from
your home to work. Different sort of cells
are coding the journey along the way.
[Rhonda]: Yeah. I’d probably would suck at
doing that, but…
[Matt]: I think, yeah, yeah. I lost that.
Yeah, some people say it’s a male aspect of
the gene, but I definitely lost it. Anyway,
but what is delightful was that you could
sort of hang, you know, a sound tone on each
one of these electrodes. And what you would
hear, and this is just, you know, for audio,
you would hear sort of “buh buh buh bump,
buh buh buh bump, buh buh buh bump,” as the
rat was running around the maze, as these
cells were learning, encoding, and creating
this kind of memory circuit essentially.
And, yeah, you know, “buh buh buh bump,” off
it goes. But what was genius is that they
kept recording. And as the rat fell asleep,
what did they hear? And it wasn’t just static
random firing, which is what we thought typically
happens during sleep. In that static of electrical
impulses at night, in a sea of electrical
noise, came out a very clear, predictive,
message, which was “brrrmp, brrrmp, brrrmp.”
It was exactly the same temporal sequence
“buh buh buh bump, buh buh buh bump, buh buh
buh bump,” but it was sped up. And we now
know that it’s during sleep that we replay,
but we replay at somewhere between 10 to 20
times the speed.
So it’s as though you’re kind of, you know,
you’ve done the, you know, recording of whatever
happened during the day, but then it gets
replayed but at times 20 or times 10. “Brrrp,
brrrp, brrrp, brrrp, brrrp.”
[Rhonda]: Wow. So is this is a long-term potentiation
part? Is this where it’s solidifying?
[Rhonda]: Now what stage of sleep is that?
I know you mentioned…
[Matt]: So that’s during deep non-REM sleep.
[Matt]: Yup, that we see that.
[Rhonda]: It’s the sleep I’m always trying
to optimize for, and it’s just so difficult.
[Matt]: And it’s difficult, you know, I think…Well,
sleep in general is difficult for so many
people, and we can speak why. So this starts
to come back to your original question about
REM sleep, though. So we get this memory replay.
It’s absolutely fascinating. We can now see
it in humans. We can even manipulate it, which
So if I teach you some information on a screen.
Let’s say it’s about a particular object,
like a fire engine, and you have to try and
learn both the object and the spatial location
of that on a screen. And then tomorrow, we’re
going to sort of come back. And I’m just going
to show you the image and say, “Do you remember
seeing it or not, yes or no?” And if you do
remember seeing it, you’ve got to place where
you thought it was on the screen. So you have
to learn the item, and you have to learn the
But here’s the great part. During the initial
learning session, before you slept, not only
do you see the fire engine but we play the
sound of the fire engine, like a fire engine
ring. And then we show you a kettle. It’s
whistling. We show you a cat. It’s meowing.
So each of these items that you’re making
has a contextual cue associated with it, a
sound. And it’s a congruent sound, fire engine
sound, fire engine, kettle sound.
And here’s the fun part. I can teach you a
hundred of these items, and then during sleep,
I’m going to replay those sounds that you
heard as if I’m trying to get into your brain
and I’m selectively reactivating each one
of those individual memories. But I’m only
going to reactivate and replay half of those
memories. I’m gonna go to replay 50 of 100
things that you’ve learned.
And then the next day, we wake you up, and
we test you. And firstly, your memory is better
after sleep. And that’s what we found. What’s
interesting is that for those items that I
replayed during sleep, they are almost twice
as superior in terms of your memory retention.
[Rhonda]: So the playing of them doesn’t disrupt
the sleep at all or…?
[Matt]: Well, that’s the key. You have to
play at a sub-threshold-awakening sound. So
we will test your auditory threshold, and
we will then play the sound at a level that
we know is below your awakening threshold.
So it’s still just enough to get in to penetrate
into the brain and tickle the memory and reactivate
[Rhonda]: That’s fascinating.
[Matt]: …but it’s not enough to wake you
up. So now, you know, you could imagine, you
know, I’ve got these science fiction ideas
of thinking, well, I learned all of this information,
and maybe I can just put my favorite playlist
on, you know, at night at low sound and, you
know, stimulate these memories. Or could it
be a study aid where you, you know, help students,
like, sci-fi stuff?
[Rhonda]: So you think coupling different
sounds maybe with, you know, learning facts,
it may actually help…
[Rhonda]: Like irrespective of replaying them,
like, during sleep, do you think just even
the coupling of the sound somehow can help,
can remember your…?
[Matt]: As long as those sounds have been
coupled and bounced to specific information.
And this is what we call context-dependent
or key-dependent memory. This is well known
in psychology over about 100 years. If you
study in the room that you’re going to take
the exam, you do better, because you actually
use cues, contextual cues from around the
room that are triggers to help you better
[Matt]: We’re just doing that with sleep.
[Rhonda]: What about caffeine? So if you study
caffeinated and you don’t take your exam caffeinated.
Is this kind of the same concept?
[Matt]: We don’t know about caffeine. It’s
an interesting thing. Caffeine may be a…it’s
a nonspecific stimulant, whether it works
with ingested substances. We know, however,
this works also with odors. So have you ever
had that experience where you’re sort of in
an airport and you’re sort of tying your shoelaces
at the security, someone walks past with that
cologne or a perfume of a person you knew,
and it instantly unlocks that memory, as if
the sound has triggered the reactivation of
the memory, and it comes flooding back.
Well, you can do the same thing with this
memory and sleep trick, where I teach you
stuff during the day, and we can puff certain
odors up your nose and associate the smell
with the learning material. And then during
sleep, you reperfuse the odor up the nose…
[Rhonda]: Oh, really?
[Matt]: …and you can get the same benefit
as well, the same doubling of the benefit.
[Rhonda]: Wow, that’s fascinating. I remember
this study that was done, that was coupling
an odor, it was like a cherry blossom odor,
with an electrical shock. So those, you know,
classical fear studies that they do in mice,
it was in rodents. And there was some sort
of epigenetic change that the breathing of
the odor, or the cherry blossom, was inducing.
That was like changing some receptor on the
glucocorticoid receptors. So it was getting
passed on to, like, the next generation. And
so even though the next generation didn’t
have the shock and the coupling of the smell,
you know, if they were exposed to the smell,
they have the fear.
[Matt]: The critical memory was translated
from one generation to the next.
[Rhonda]: Yeah. So, anyways, it’d be fascinating.
You don’t know what role sleep plays in the
[Matt]: Right. So if we all start doing this,
you know, are our children going to be incredible
memory replayers, you know, by the way of…But
to come back to your…I’m sorry I’ve taken
a desperately long time to answer your original
question, which was what about REM sleep.
So what we’ve spoken about is the first two
of the three stages of memory processing with
sleep, sleep before to get the brain ready
to lay down memories, sleep after to grab
a hold of those individual memories and cement
them into the neural architecture of the brain.
Once you’ve done that, though, there’s a final
step. And that seemed to not depend on deep
non-REM sleep but instead depend on rapid
eye movement sleep or REM sleep, which is
what most of us know as dream sleep.
And it’s during dream sleep that your brain
essentially performs informational alchemy,
is what I would describe it as. It’s a little
bit like group therapy for memories that sleep
has gathered in all of the information during
the day. And during non-REM sleep, which always
comes first, by the way, in our sleep cycle.
We always have non-REM sleep first, then REM
sleep second, then non-REM sleep again, then
REM sleep second.
And we don’t know why there is no good explanatory
data suggesting why non-REM sleep always comes
first and REM sleep comes second. But I’ve
put forward the theory that, for information
processing, it makes sense. Which is that
non-REM sleep first to just get what you’ve
learned and lock it into the brain. REM sleep
then comes along, and REM sleep starts to
fuse all of the information that you’ve recently
learned with the entire back catalog of information
that you’ve got stored up across a lifetime
And it’s this sort of…Essentially, REM sleep
is creating a revised mind wide web of associations.
And I’d like to sort of think of what’s going
on with REM sleep. And we’ve done lots of
these studies to look at this and these clever
ways that you can look at sleep and associative
memory processing and building new novel connections.
And it’s almost like memory pinball, where
you take these new memories and you sort of
launch them up, and you start bouncing them
around into the architecture of information
within the brain. And you’re starting to test
associations. You’re starting to say, you
know, “Should this new information be connected
to this? Maybe not. Should it be connected
to this? Maybe not.”
Now some of that happens whilst were awake
during the day. We make obvious connections.
But what’s strange is that we make connections
during REM sleep but they’re not of the same
kind. The connections that we’re making during
REM sleep are the longshots. This is the bizarre…
[Rhonda]: The bizarre, right.
[Matt]: …strange. You know, it’s sort of…it
would be the equivalent of saying, during
the day, we take this information, and the
connections we make are like a Google search
gone right, which is the first page is all
of the things that are most related. And it’s
very obvious. Page 1, that’s directly related
to what I inputted.
During REM sleep, it’s almost as though you
input the search term and you’re immediately
taken to page 20 of the Google search, which
is about some field hockey game in Utah. And
you think, I don’t understand. Oh, that’s
interesting. I see what you’re talking about.
So we make these bizarre leaps of associative
memory processing faith during REM sleep.
And that’s why we now understand that it’s
REM sleep that helps us divine remarkable
creative insights into previously impenetrable
And you can see this throughout the history
of human beings, this dream-inspired insight,
scientific demonstrations. You know, August
Kekulé divined the idea of a benzene ring,
these double carbon rings by dreaming of a
serpent that swallowed its tail. Dmitri Mendeleev,
you know, came up with the Periodic Table
of Elements by way of dream inspiration. And,
you know, people have won Nobel Prizes, Otto
Loewi won the Nobel Prize for the demonstration
of chemical transmission across nerve cells.
And he dreamt of the experiment that helped
improve that. He didn’t dream of the concept
itself, but he dreamt of the experiment to
Wonderful, artistic demonstrations of this,
too. You know, Paul McCartney has written
innumerable songs, it turns out, by way of
his dreams. Keith Richards came up with the
opening chords of Satisfaction by way of dream-inspired
So REM sleep takes that third component of
information processing. And I think it’s what
defines us differentially from computers in
part, which is that deep sleep is about knowledge,
which is gathering all of the information
and holding on to it. REM sleep, I would argue,
is about wisdom, which is knowing what it
all means when you fit it together, you know.
That’s what I want from a good student. Don’t
just give me dry-book learning. Do you really
understand it? Can you apply it? Are you creative?
That’s dream sleep.
[Rhonda]: Right, a deep thinker. Why do you
sometimes not remember your dreams and sometimes
you do remember your dreams? Do you have any
[Matt]: Yeah. So some of it seems to be about,
if you wake up out of that dream sleep period
and then you go back into sleep, the awakening
can sometimes help you commit that experience
[Rhonda]: Oh, okay.
[Matt]: But there are people who say that
I never remember my dreams, you know. We can
bring those people into my sleep center, and
we can, you know, wake them up in the middle
of dream sleep, and they’ll say, “It’s remarkable.
For the first time, I was dreaming.” And the
answer is no. It’s not the first time that
you are dreaming. It’s just the first time
that you’ve actually remembered a dream, because
it’s the first time you’ve typically woken
[Rhonda]: Okay. My mother-in-law, you know,
claims that she doesn’t dream. And, of course,
I’m like, “No, you have to dream.”
[Rhonda]: That must be…
[Matt]: There are a selection of patients
that have a lesion in a part of the prefrontal
cortex, in their white matter, which are these
big sort of informational fiber tracts that
communicate impulses. If you get a lesion
deep down there, we do seem to genuinely see
a cessation of dreaming in those patients.
By the way, I would…I didn’t even feel confident
to write this in the book, and it’s still
a theory that I’ve never really heard in public,
but go with me on this. Which is, I think
that we may actually remember all of our dreams,
or it’s possible that there’s a tenable theory.
The problem is we don’t have access to those
dreams. Those dreams are memorized, and they
are available. They’re just not accessible.
I think what happens as we wake up is that
we lose the IP address to those memories.
And the reason I believe this to be potentially
true is have you ever had the experience where
you wake up and you think, I was dreaming,
and I know I was dreaming. And you try as
hard as you can. The harder you try, the…
[Matt]: …worse the memory-recall goes? And
then you think, forget it. Two days later,
you’re walking along, and you see a street
sign, and all of a sudden, it triggers the
unlocking of that dream memory. You think,
oh, that’s what the dream was about.
As a neuroscientist, that tells me that the
memory was present. The memory was available.
The problem was accessibility. You couldn’t
gain recall access. So the information is
there. It’s just not accessible. If that’s
[Rhonda]: It’s happened to me just even after
I’ve, you know, when I go to bed. You know,
later that night, I hit the pillow, and all
of a sudden, I remember the dream right as
I’m hitting the pillow.
[Rhonda]: That’s happened to me more than
once. But you’re right.
[Matt]: Right. It sort of tells you that there
is…it’s almost a scary prospect, which is
that maybe every single one of our dreams
throughout life are stored and are present
and determine our behavior to some degree,
because we know that there is an enormous
amount of information that changes our behavior
or decisions that goes on below the radar
of consciousness, implicit memory. That could
be true for dreaming, too.
And I think I’ve got an experiment that we
may be able to design to actually get at this.
And if that’s true, it should, hopefully,
radically change our view of dreaming, that
dreams are ephemeral, that they dissolve quickly,
they’re forgotten, and they don’t influence
us as human species.
[Rhonda]: That would be pretty groundbreaking.
[Matt]: Yeah, we’ll see.
[Rhonda]: You just had a study that I just
read, I think, yesterday on sleep and it affecting
behavior, loneliness, or…
[Matt]: Yeah, so we just published a sleep
study demonstrating that sleep loss will trigger
viral loneliness. And it was a three-part
study. I mean, firstly, the reason that I
got into this was loneliness is a killer.
We know that there is a massive epidemic of
loneliness now in industrialized nations.
Being lonely increases your mortality risk
by about 45%. In other words, being lonely
is twice as risky for your death concern than
[Matt]: …which is striking.
[Rhonda]: Yeah. There was actually a study
showing loneliness changes, like, a massive
amount of gene expression and, like, up-regulates
NF-kappaB, cortisol, like all these pro-inflammatory
genes. So it makes sense that it’d be associated
[Matt]: And what’s bizarre about loneliness…By
the way, I’m taking a complete…this has
got nothing to do with sleep. But if you look
at the profile of your gene expression and
your immune system, you’ve got some immune
components that will go after viruses. And
viruses can only be transmitted from one human
being to another by way of touch. They can’t
live outside of our bodies. Bacteria, so if
you scrape yourself on a fence, like, you
know, walking past it, you can get a bacterial
infection because bacteria can live outside
of the body.
When you become lonely, your gene expression
shifts you away from a profile of immunity
that normally deals with viruses and pushes
you to more towards a bacterial defense profile.
[Matt]: Isn’t that incredible?
[Rhonda]: Yeah, you should send me that study.
[Matt]: That your psychology…
[Rhonda]: Yeah, that’s fascinating.
[Matt]: And there’s a couple of folks at UCLA,
who, if you ever have interest in this area
of how loneliness, the mind, the mood…
[Rhonda]: Oh, totally. Yeah.
[Matt]: Okay, I’ve got to give you these people.
I’m a complete fan of…
[Rhonda]: Please do.
[Matt]: …their work. And they did this study.
And it just blew my mind. How could a concept
that is so sort of, you know, out there, and
some people almost don’t, you know, believe
in loneliness. Toughen up. What’s wrong with
you? Go out make some…How could that change
the expression of your genes and even alter
how you, the organism, fend for yourself from
an immunological perspective shifting you
from viral to bacterial defense. But, anyway…
[Matt]: So coming back, I was desperately
concerned about the state of loneliness. What’s
interesting, I was reading a lot at the time,
because we do a lot of work with sleep and
psychiatric disorders including anxiety, and
when I was reading the studies where they
would take animals and they would deprive
them of sleep, you’ve got this anxiogenic
profile where you got cortisol increasing,
you got shift in insulin and glucose regulation,
all of the bad things that you would not wish
to happen, an anxiety increase, they had fear-like
behavior all by way of just sleep restriction.
But what was also interesting is that sometimes
the researchers would know, despite not measuring
it systematically, that the animals, who would
often be secluded by themselves in the cage,
even when they were with other conspecifics,
and other conspecifics would not approach
[Matt]: And so it seemed to me just from reading
this, I thought, well, this seems like an
animal phenotype of human loneliness. And
it seems to be caused by a lack of sleep.
So we decided we had to do the study.
So the first part of the study, we took a
group of individuals, and they went through
the study twice. They were either deprived
of sleep for an entire night, or they got
a full eight hours of sleep. The first test
was do you have a social repulsion boundary,
and that boundary is increased when you are
sleep-deprived. So I think everyone has that
sense that if I start moving closer to you,
you think, okay, do you know what, at some
point, that’s kind of enough. That’s about
my close distance.
What’s interesting is that if I ask a sleep-deprived
an individual to stay put, and I ask you,
as an experimenter, to walk towards the sleep
deprived individual, and the individual says,
“Stop,” when they feel comfortable relative
to when that very same individual has had
a full eight-hour night of sleep. When you’re
sleep-deprived, you decide to push people
a further distance away from you. So you have
a lower desire for social proximity and social
Second, we then replicated that finding, but
now we had them inside the MRI scanner. Because
we wanted to see what was changing the brain
to produce this social repulsion. What we
found was that the regions of the brain that
are essentially an alarm network, which is
a sort of a stay-away-from-me network that
is sort of in the parietal cortex and the
premotor cortex, it’s sort of what we call
the near space network. So it creates your
comfort of boundary. And when objects start
to approach you, it alarms to say, “Incoming.
Be cautious. Be wary.” That part of the brain
became hyperactive when people were sleep-deprived…
[Matt]: as if you were getting this repulsion
signal from the brain. If that wasn’t bad
enough, the other parts of the brain that
have been called the theory-of-mind network,
which sort of helps you understand the intent
of other people, it’s a pro-social network
in the brain, it cooperates pro-social interaction,
that part of the brain was shut down by sleep
deprivation. So it’s a double-edged sort of
So we weren’t satisfied with that. Next, we
wanted to say, “Could someone who just looked
at these sleep-deprived individuals, could
they actually judge them as being lonelier
and looking lonelier and beaten, sort of perceived
as lonely, even though they knew nothing about
So in the experiment with the sleep-deprived
individuals, we also did videotaped interviews
with them. And we just asked them general
questions. Tell us about a movie that you
watched? Or what was happening in the news
this week? Just bland stuff. And then we got
1,000, I think was it was over 1,000 people,
1,083 people online. And they knew nothing
about the experiment. They didn’t know it’s
about sleep, sleep deprivation, knew nothing.
And we showed them just a 60-second clip of
these people when they’d have a good night
of sleep and when they were sleep-deprived.
And we just asked them, “How lonely does this
person appear to you?” And they knew nothing.
But despite knowing nothing, they consistently
and reliably rated the sleep-deprived version
of the individual as seeming lonelier.
We also asked them, “Would you socially interact
with this person? Would you friend them on
Facebook? Would you work with them in a business
environment?” And they consistently rated
that they would prefer not to engage and interact
[Rhonda]: Is that because they just looked
unhappy or looked…?
[Matt]: Well, we actually think it’s a collection
of things. It’s that their appearance, but
also their vocal tone, is very different.
We think there’s a lot of…This is now one
of the key things. What’s communicating this
[Rhonda]: Oh, so they are listening to them
[Matt]: So they watched them, and they listen
to them speak. So they could hear them as
well. So we demonstrated that. There was,
unfortunately, the social repulsion on both
sides of the equation. When you’re sleep-deprived,
you yourself don’t want to have anything to
do with other people. And that perhaps wouldn’t
be so bad if people would only at least come
to your rescue, because they would see you
in need. The opposite is true. People find
you socially repulsive as a consequence. So
there’s a push from both sides of the social
The next thing, we asked those people who
were rating the sleep-deprived individuals,
we also said, “Look, how lonely do you feel
after just the 60-second clip?” And they themselves
felt lonelier after interacting with sleep-deprived
individuals. In other words, this contagion
of sleep-deprivation-induced loneliness.
[Rhonda]: So I wonder how much of this can
be translated to, like, someone that, say,
for example, gets only five or six hours of
sleep versus, of course, not getting a full
night sleep. You know, maybe there’s, like,
a little, just a little bit of this penetrating…
[Matt]: We then ask that question. That was
the final part of the study, which is that
we said, “Okay, this is extreme sleep deprivation,
and most of the population does not undergo
this. What about a more ecological version?”
So we tracked hundreds of people across two
nights of sleep. And we asked, just by a subtle
variation of nature, our small perturbations
of sleep from one night to the next, do they
predict how lonely you experience yourself
to be from one day to the next? And these
are small minute changes in sleep efficiency,
just small reductions in sleep of tens of
Lo and behold, even just that small change
in your sleep from one night to the next,
we could measure, predicted how lonely you
would experience life the next day from one
day to the next. So it doesn’t even take,
you know, two hours of sleep reduction.
[Matt]: Small minutes.
[Rhonda]: I remember reading somewhere, too,
that isn’t the, like, amygdala, like hyperactive
or something happens, there’s not an inhibitory
signal that occurs if you’re sleep-deprived
[Matt]: That’s right.
[Rhonda]: Is that correct?
[Rhonda]: So then you’re feeling more…you’re,
like, alarmed and, you know, just anxious
and feeling that in a, say, threat…
[Matt]: That’s right.
[Rhonda]: …that ongoing threat that really
[Matt]: Exactly, yeah. So we published the
study in 2007 where we, again, sort of sleep-deprived
people, put them inside an MRI scanner, and
we showed them increasingly negative and aversive
and unpleasant images. And what we saw is
that, relative to people who’d got a full
night of sleep, the amygdala, this sort of
emotional epicenter for the generation of
strong, emotional, impulsive reactions, that
deep emotional center was 60% more reactive
under conditions of a lack of sleep.
And then we asked why. Why is your emotional
brain so sort of sensitive and erupting with
such extraordinary activity? And what we then
went to find, or went on to find out in later
work, was that another part of your brain
called the prefrontal cortex that sits directly
above your eyes here, and particularly the
middle part right between your eyes, that
part of the brain acts almost like the CEO
of the brain, of your emotions, and your hedonic
impulses. And it sends sort of an inhibitory
top-down regulatory control. It’s sort of,
like, the brakes on the gas pedal of your
That part of the brain was shut down by sleep
deprivation, and you’d lost that communication
to the amygdala. So now you, from an emotional
standpoint, you were all emotional gas pedal
and too little regulatory control brake as
[Rhonda]: It’s really interesting. This work
kind of reminds me of…I’m not sure if you’re
aware of any of this research. A lot of it
has been done by Dr. Molly Crockett, who,
I believe, now she’s at Harvard. But she has
done a lot of studies looking at serotonin
depletion in the brain. And basically, you
can induce that by giving acute tryptophan
depletion, giving someone like branched-chain
amino acids to compete with transport for
tryptophan in the brain, which then basically
drops serotonin levels. I mean, you can drop
your serotonin levels down to, like, 10%.
[Matt]: And mood is…
[Rhonda]: And the same thing happens where,
exactly what you were describing, the inhibitory
signal that happens from the prefrontal cortex
onto the amygdala is, like, stops. And so
people become extremely impulsive. Terrible
moods, a little more aggressive, their long-term
planning shuts down, and they just, like,
going for the short-term gratification. Very
similar. So it’d be kind of interesting…I
don’t know how serotonin would be related
to all that, but there must be some sort of
[Matt]: Yeah. I mean, and I think there’s
a number of different, I think, neuro chemicals
that can produce that same kind of neural
phenotype and as it were. But what struck
me was that when I looked at that neural signature
of sleep deprivation for the emotional brain,
it was not dissimilar to numerous psychiatric
conditions. And that then now, gosh, 11 years
ago, I’m showing my age, but that set sort
of, you know, the sleep center off on a completely
new trajectory of work. And we’re doing a
lot of this work in sleep and psychiatric
And I think one of the most fundamental things
that I can say at this point is that we have
not been able to discover a single psychiatric
condition in which sleep is normal. And so
I think sleep has a profound story to tell
in our understanding, maybe our treatment,
I don’t know about prevention, but possibly,
of grave mental illness. And psychiatry has
known this, by the way, for, you know, 40
or 50 years. It’s always been documented that
sleep disturbance goes hand-in-hand with psychiatric
[Rhonda]: Maybe there’s some sort of complex
gene environment interaction. A few people
that are more genetically susceptible and
are getting…losing the sleep, or, like,
the ones that are kind of pushed into a disease
[Matt]: And we’ve seen this, too, that if
you look at that emotional brain reaction
signature that I just sort of described, and
you repeat that, but with people who are high-anxious
and low-anxious, and we know some of the genes
that are associated with being high-anxious
and low-anxious. So we’re using anxiety as
a sort of a proxy for perhaps a particular
genotype here. What you see is that it’s those
high-anxious people who are the most vulnerable
to this impact of a lack of sleep. Those who
are low-anxious still have a bad outcome,
but it’s nowhere near as bad. So there seems
to be sort of interactions here between sleep
loss and your basic trait levels of being
sort of a nervous, anxious type to begin with.
[Rhonda]: That makes sense.
[Matt]: And those are the people who are,
sadly, the people who typically don’t get
a good night of sleep anyway.
[Rhonda]: Right. So you’re saying anxiety
is, like, one of the things that stops me
[Matt]: It’s the principal trigger of insomnia.
[Rhonda]: Yeah, really, true.
[Matt]: It’s the model of insomnia right now,
is that you get…And if you look at the nervous
system, that’s how we understand insomnia
right now, is that its principle is…I think
ultimately we’ll find that there are multiple
flavors of insomnia, different forms. We already
categorized two of them. We’ve got what we
call sleep onset insomnia and sleep maintenance
insomnia, difficulty falling asleep, difficulty
staying asleep. They’re not mutually exclusive.
You can have both, or you can just have one
or the other.
But coming back to it, I think the overarching
biological red thread narrative of insomnia
is an amplified fight-or-flight nervous system,
that your nervous system is split into these
two branches, what we call sort of the sympathetic
and parasympathetic parts of your autonomic
nervous system. The sympathetic is anything
but sympathetic. It’s poorly named. It’s the
fight-or-flight branch of your nervous system.
It ramps you up, charges you up, releases
You constantly see an overactive, sympathetic
nervous system in people with insomnia. And
when you measure their cortisol across the
24-hour period, in most of us, just as we’re
getting to our natural bedtime, cortisol just
starts to now drop down. We start to see that
cycling down of cortisol. The opposite happens
in people with insomnia. You get a continued
rise right around that bedroom period. And
it seems to be very predictive of sleep onset
If you look throughout the night, cortisol
then starts to plummet, and it drops beautifully
down. It’s part of the reason why deep sleep
is the best form of natural blood pressure
medication that you could ever wish for. Your
heart rate drops down, your vessels relax,
cortisol drops down. But in other insomnia
patients, we see this bizarre spike in cortisol
in the middle of the night. And it predicts
nighttime awakenings. It predicts sleep maintenance
[Rhonda]: I have experienced…So that’s one
of the problems that I actually have. It’s
much, much better now that my stress level
is, like, maintained. At graduate school,
I would get nighttime awakenings where my
heart would start racing. And I would wake
up thinking that there was a spider or some
kind of threat. And I would scream, and sometimes
fly out the bed. I mean, you know, and just…it
would scare my husband, you know. At the time,
we weren’t married. But, I mean, you know,
these nighttime awakenings, it was something
that’s dated back for quite some time. But
really, it manifested during a very stressful
period, and that was graduate school.
[Matt]: We see that…
[Rhonda]: Got much better.
[Matt]: …so frequently.
[Matt]: But if you can think about that as
sort of, you know, a stress management component
to insomnia, you know, it’s part of what we
call cognitive behavioral therapy for insomnia,
which is sort of dealing with that stress,
you know, meditation. There’s great apps out
there, like Headspace, for example.
And the data on meditation and insomnia is
very very powerful. You know, I’d known about
it a little bit, but I hadn’t read really
all of the studies until I started researching
it for the book. And I was so convinced that
I started meditating. And I haven’t stopped
since. Because it was, you know…Typically,
I’m not a bad sleeper. I’m a light sleeper.
I’m a pretty good sleeper. I found it hugely
useful for times when I was under stress,
or when I was traveling and jetlag. It’s very
But that underlying theme, I think, as a message
for insomnia, it’s not the only cause of insomnia,
but it seems to be if there’s one common sort
of rule that binds many of the patients with
insomnia together, it’s this overactive fight-or-flight
branch of the nervous system. And if you can
settle that down, you are certainly on the
path towards better sleep.
[Rhonda]: Right. And to kind of just…Another
point that you made talking about, you know,
the hormonal response and the cortisol rising
typically when it’s supposed to be falling,
that kind of prevents you from falling asleep,
there is some interesting research that I’ve
read where, and I know you and I have talked
about how the importance of bright light exposure…Bright
light exposure for six hours a day…I mean,
no one does that nowadays. We’re always inside.
So it’s rare unless you’re, like, working
out in a nature park or something.
[Matt]: The irony of these things…
[Rhonda]: Right, you know.
[Matt]: Yeah. It’s for the camera. I promise.
[Rhonda]: Right. But, actually, it was shown
to lower cortisol by 25%.
[Rhonda]: So this is, like, you know, another
kind of I don’t know if that would even help
someone with the anxiety or, you know…
[Matt]: No, I think it’s…there’s no studies
testing it yet, but there are studies…So
just sort of to go back to make this point,
we normally have a circadian rhythm, this
beautiful sort of 24-hour rhythm. And we human
beings were diurnal, and we like to sleep
at night, be awake during the day. We have
this awesome upswing of our circadian rhythm
sort of once we wake up sort of 7, 8, 9, 10,
11. It starts to peak during the day. It drops
down a little bit in the mid-afternoon. And
that’s why you sort of get around meeting
tables in the middle of the afternoon these
sort of, you know, head nods. It’s not people
listening to a good music. It’s like, actually,
it’s a pre-programmed dip in your alertness.
And then it rises back up, and then it drops
down at night.
And one of the ways that you can get this
sort of what you would want, which is a nice
sinusoidal wave, you want a nice strong peak
of the circadian rhythm during the day so
that you’re awake and you’re active during
the day and you’re productive, and then you
want an awesome sort of trough throughout
the night so that you sleep soundly, deeply,
and in a stable fashion.
And the way that you can sort of help your
circadian rhythm have that wonderful peak
and delightful trough is by getting lots of
daylight during the day but lots of darkness
during the night. And we are a dark-deprived
society in this modern era. And it is a huge
problem in the evening.
But I think people have underestimated that
we are a light-deprived society during the
day. So what happens is that your brain goes
through life in this kind of almost stupor
state where it’s not getting enough daylight
to really keep it ramped up throughout the
day so you’re sleepy throughout the day and
you’re tired, but then we’ve got too much
light in the evening so you end up being awake
at night, and then you’re sleepy during the
day, you’re awake at night.
And so, you know, almost like, I’m gonna call
it a seesaw. I think we call it teeter-totter
here. You know, during the day, you want daylight
to come in and force you all the way on to
the on switch and you’re active and awake.
And then at night, you want the signal of
darkness to come in to trigger the release
of a hormone called melatonin to shift you
all the way into the off position so you go
into deep sleep and a sound sleep.
But now, with artificial light and staying
out of bright light, so the teeter-totter
has just pushed a little bit to one side,
and then sort of with not enough darkness
at night, it’s only pushed a little bit down
on the other side. So you kind of have this
flip-flop switch that…it’s like a dimmer
switch that is basically just on dim for 24
hours rather than light and complete darkness,
if that makes some sense.
[Matt]: It’s terrible analogy, but…
[Rhonda]: Do you have any idea, like, how
much bright light exposure, like, you know…Let’s
say, you know, a lot of people work in, like,
a little cubicle where they could don’t even
have windows. I mean, they’re just like in
the middle of this, like, little rooms with
no actual sunlight coming in, you know, is
something like if you wake up in the morning
and you go outside for 30 minutes or an hour
first thing in the morning, like…Most people
are drinking their coffee. Well, maybe what
you need is bright light exposure instead
of your coffee, or drink your coffee outside.
[Matt]: Right, yeah. Take the coffee on…
[Rhonda]: Yeah. Do you know how much light
you need? And if you don’t have light outside,
let’s say you live in London, and you’re in
the winter, and there’s no…It’s just gray.
[Matt]: Even that, even on a cloudy day, the
lux intensity of light far exceeds that that
you would have from incandescent light or
at sort of typical lights inside of a building.
[Rhonda]: Okay. So now my second question,
the lux amount, so do you know?
[Matt]: Yeah. So I think, I mean, if you look
at the studies, once you get over a sort of,
you know, about 5,000 to 10,000 lux, you can
have a pretty powerful effect. I don’t believe
there’s…and I could be wrong. And Satchin
Panda, our good friend, who is just A plus…In
fact, yeah, if you’re watching this, like,
stop watching this now. Just go and watch
the Satchin Panda podcast. He’s much more
powerful and eloquent than I am. But they
have looked at the degree of exposure to outside
light, not necessarily the intensity of light
during that outside exposure time.
So I don’t think we yet understand exactly
what the dose response is in terms of lux
intensity. What we do know is that getting
30 to 40 minutes of outside morning light
is critical. But here’s the trick. Here, in
California a lot of people make this mistake,
but even in London, it happens despite cloudy
day, people put shades on in the morning.
Don’t do that. I know it looks good. But don’t.
Let that natural light penetrate your eye.
There’s a retinal mechanism that goes through
to your thalamus that then goes through the
hypothalamus that regulates your circadian
clock. You need that light penetration. You’re
losing all of that good stuff if, or some
of it, if you put shades on.
Wear some protection, that’s fine. Just nix
the shades in the morning. In the afternoon,
reverse the trick. And this is actually a
very good tip for jet lag. Jet lag is essentially
an extreme form of what most of us have, which
what we’ve been describing this diluted amount
of light during the day and then too much
light at night. Jet lag, you really should
get out in the morning, 40 minutes of daylight,
no shades. And in the afternoon, it’s fine
to go out. But when you go out, now is the
time to put shades on, because you can start
to encourage even then the release of melatonin,
which is that hormone of darkness, which signals
the timing of healthy sleep.
[Rhonda]: So about what time in the afternoon
would you say shades are…?
[Matt]: So I would say, you know, probably…it
depends on your bedtime, if you’re a morning
owl or an evening lark, and we can speak about
chronotypes. So it really depends on when
you’re planning on going to bed. But let’s
say that you’re planning on going to bed at
about 10 p.m., I would say, if you’re sort
of going out after about to 4:30-ish, now
is a good time to maybe start to help dilute
down some of that light.
But then, you know, in the evenings, you know,
we are so bathed and saturated in light. And
yes, we can speak a lot about LED screens,
and they are impactful, and there’s been lots
of work on that, some of which haven’t replicated
but many of which have. I think the bigger
problem is just overhead lighting in general.
We’re just infused by, in every room that
And my recommendation has now been, in the
last hour before bed, just turn off half of
the lights in your house, you know. We don’t
necessarily need all of them blazing in the
last hour before bed. And when you do that,
it’s quite surprising how soporific and somnogenic
it actually is, you know. And I want to do
the experiment, although someone beat me to
the experiment, and they did it…you know,
it’s one of those studies when I read it,
I just thought, my first reaction, I’m not
a big person. My first reaction was, oh, I’m
[Rhonda]: Oh, yeah.
[Matt]: I was like, oh, I wish I’ve done…And
then I just thought, this is a brilliant paper.
I can’t wait to teach it. They took a group
of people. They looked at their habitual amount
of sleep that they would typically get. And
these people are getting sort of seven and
a half to eight hours. And they would ask,
you know, “When are you gointo sleep?” And
sort of most of them would go to bed like
11 and sleep through till 7.
And then they took them out of that typical,
you know, modernity environment, and they
took them out to the Rockies. And they had
sleep tracking equipment on them. They took
them there for several weeks. And there was
no electricity whatsoever, not even a torch,
not even a headlamp from a car, nothing. And
then they look to see what changed.
The first thing was that these people went
from sleeping, you know, an acclaimed seven
and a half or seven hours of sleep that was
their norm. It was actually just below seven,
saying, “That was fine. That’s all I needed.”
To then, actually, when they had no watch,
they didn’t know when to wake up, no alarm
clocks, they ended up sleeping closer to nine
hours a night, which is what we typically
see when you saturate sort of people away
from or dislocate them from modernity.
[Rhonda]: So would you say that’s a good sleep
[Matt]: Well, I think somewhere between seven
to nine is what we recommend. But I think
when you do this in a healthy, young people,
and these were healthy, young people, they
seemed to acclimate to a sleep amount that
was somewhere between sort of eight to nine
hours of sleep. So I think it’s good evidence
that, you know…You can look at how hunter-gatherer
tribes were sleeping. And we’ve studied, you
know, these people. And they actually sleep
in a strange manner. We can get back to that.
And people have tried to use them as the gold
standard as to how we should be sleeping.
[Rhonda]: They’re a lot more active and…I
mean, they’re so different, right?
[Matt]: I don’t think it’s a good control.
I think we should say, “Let’s take modern
human beings and let’s just take them out
of all context of modernity. And let’s see
how they’re sleeping. Let’s just sort of put
them on an ad-lib buffet of sleep. And they
can just sleep as much as they want. They’re
not told when to wake up and sort of when
to go to bed.” And they seem to sleep what
we now think of as a natural amount, which
is somewhere between, if you look at the distribution,
seven to nine hours.
[Rhonda]: And these are younger individuals?
[Matt]: Yup. And we can speak about, and I
hope we speak about sleep and aging. But what
was also interesting is when they slept, not
just how much they slept. They started to
go to bed earlier and earlier and earlier,
and they started to wake up a little bit earlier
and earlier. And the total duration of sleep
expended. But where that expanded amount of
sleep was positioned on the 24-hour clock
was dragged back because they weren’t influenced
by these cues of, you know, too much daylight
at night. Temperature is another one that
I’d love to speak about, too.
But what’s fascinating is that when you look
at hunter-gatherer tribes, all these experiments
of sort of true nature, the natural point
of middle point of sleep, the middle phase
sort of time of that eight to nine hours sleep
phase, came somewhere between midnight and
1:00 p.m. And I often ask people this question,
you know, “Have you ever thought about what
the term midnight actually means?” You know,
it means the middle of the solar night, which
is the time when most of us should be in the
middle of our sleep phase.
But now, in the 21st century, we’ve gone through
the, you know, the agrarian sort of, you know,
pushed into the industrial era, and now into
the digital era. Now midnight is the time
when we maybe check Facebook for the last
time or think about sending that last email.
So not only has the duration of our sleep
decreased through the influence of the modern
times but also when we’re sleeping has been
dramatically shifted, too.
[Rhonda]: Right. I know I’ve made some changes
a few years ago to my place, where now, I
was telling you, I have Philips hue lights
that turn on red light. And they come on actually
quite early. They come on around 5 p.m. In
fact, when we have visitors, they start to
go crazy when sunset, and it’s like red, and
they’re like getting sleepy, you know. It’s
like, “Why is it so dark in here? Can we turn
the lights on?” And it’s like, “No, because
that’s what you should…You’re supposed to
be getting the sleepy right now. You’re supposed
to be…It’s 6:00, 6:30.” Well, depending
on what time of year it is, you know, “The
sun is setting. You should be getting sleepy.”
[Matt]: Yeah. It’s one of the few triggers.
[Rhonda]: Yeah, it really works.
[Matt]: Temperature is the other key trigger
that…I think that’s probably, you know,
in the past three years, that’s what we’ve
really understood, is that temperature is
as powerful a trigger of sleep organization
and sleep depth as light is.
[Rhonda]: And sleep depth.
[Matt]: And sleep depth as well. So what typically
happens is that, for you to fall asleep and
stay asleep, your body needs to drop its core
temperature by about 1 degree Celsius or about
sort of 2 to 3 degrees Fahrenheit. And that’s
the reason that you will always find it easier
to fall asleep in a room that’s too cold than
too hot, because the room that’s too cold
is at least taking you in the right thermal
direction for good sleep. It’s dragging your
body down into a cooler realm.
And when you look at these hunter-gatherer
tribes, you know, the San in Namibia, you
can look at them, whose way of life hasn’t
changed for thousands of years, they don’t
go to bed and necessarily as immediately as
the sun goes down. They usually go to bed
maybe sort of 8 or 9 in the evening, several
hours after sundown.
But when they do start to go to bed is when
the temperature drops. As the sun drops, the
ambient temperature remains for a little while,
and then it starts to drop, too, as the land
cools, and the ambient temperature cools with
the land because darkness has ensued. That
seems to be a thermal trigger for them getting
sleepy and falling asleep.
Then, when you look at when they wake up,
they typically wake up 15 to 20 minutes before
dawn. So it’s not light that seems to be necessarily
the trigger instigating the awakening. It’s
actually the rise of temperature.
[Rhonda]: And that’s on a circadian rhythm,
[Matt]: And that’s on a circadian…So what
is training us to our natural sleep rhythms
is both temperature and light. So this comes
back to our modern homes, where we go into
offices, and we don’t necessarily have the
rising warmth of the day to activate us because
we’re set at 70 degrees, and then we go home,
and our thermostats are set at 70 degrees
again or whatever your standard temperature
is. And we don’t get the thermal cue through
our bodies to say, “It’s time for sleep.”
So no wonder our sleep is worse.
And if you manipulate people’s core temperature,
and you can do this through showers, hot baths,
you can do it through saunas, but there’s
a great experiment where they essentially
designed what it looked like a kind of a wetsuit.
And throughout the wetsuit were all of these
plastic veins running through them. And the
researchers were able to exquisitely control,
through water at different temperatures, the
temperature of any part of your body. So would
I want to warm your hands and your feet? Or
would I want to warm your core?
And so what they were able to do is…These
people I mean talk about being stripped of
your dignity, you know. You get into this
like wetsuit, you zip it up, and you say,
“Right, I’m off to bed.” In you go. And then
they would manipulate core body temperature
exquisitely. And lo and behold, when they
dropped the core body temperature, they were
able to induce sleep quicker, and the amount
of deep sleep, what we call slow-wave sleep,
which is deep non-REM sleep or Stages 3 & 4,
also called slow-wave sleep because of these
big powerful slow brainwaves, that deep, rich
non-REM sleep was increased somewhere between
about 10% to 20%, which is non-trivial.
[Matt]: Then they said, “Well,” this is in
sort of, you know, young healthy people, “what
about people with insomnia, and people who
are older.” Because older people struggle
with sleep and, of course, insomnia. And they
were able to get even greater of mileage out
of the thermal manipulation with those cohorts,
too. So I think it’s a very interesting intervention
possibility to try and augment and manipulate
[Rhonda]: So do you think those things, like
the chili pads that are out there, you think
those can potentially help deep sleep?
[Matt]: Yeah. I don’t know of any data from
them yet. All I know is the experimental data
that we’ve looked at, you know, with manipulation
of body temperature. They did this in rats
awhile back, by the way, which is they would
warm their paws. And when they warm their
paws, the rats fell asleep quicker and stayed
And you think, hang on a second, you just
told me that you need to cool the body, but
you’re talking about warming them up. And
this is the reason why probably saunas work,
hot baths work, and showers work for the opposite
reason that most people believe them to work.
You think, you know, I get out of a hot bath,
and I’m nice and toasty, I get into bed, and
I fall asleep faster.
That’s not true. What happens is that you
get into the bath, you get massive vasodilation.
All of the vessels open up on the surface
of your skin. That draws, it almost charms,
the blood out from the core of your body to
the surface. And your skin and your hands
and your feet especially act like these wonderful
thermal radiators. And they dissipate the
heat. So you get out of the bath, and your
core body temperature actually plummets, and
that’s what you need for good sleep.
And I suspect this the reason that saunas…I
mean, I think you’ve had experience it too.
[Rhonda]: Yeah, definitely. It definitely
affects my sleep. And going from a sauna into,
like, an ice cold shower and then doing kind
of like going back and forth really really
helps sleep. And I’ve used that also going
to other countries. Getting in the sauna and
going the cold shower and then back in the
[Matt]: And triggering it back.
[Rhonda]: It helps with my jet lag.
[Rhonda]: I mean, it’s an end of too, because
my husband experience the same thing. But,
you know, there’s definitely something…there’s
something there. And what you’re saying absolutely
happens. The sauna increases vasodilation,
blood flow goes to the skin, and so dissipation
of heat. That all makes perfect sense. So
[Matt]: I’m just gonna, very quickly, just
grab a quick mouthful of water.
[Rhonda]: The hot baths and hot showers are
good before bed.
[Matt]: Yeah. And I’ll just mention, I think
one thing that’s I think fascinated me, though,
about the benefit of exercise…so exercise
also has a really nice, powerful benefit on
sleep, although it’s a two-way street, and
remind me to talk about that, in a good way.
But something else has occurred to me with
saunas, too. And I was thinking about this,
because I think I know that you’re a fan.
I think you’ve mentioned this before. One
of the ways that you can induce sleep is that
you can increase a lot of the immune factors,
things like cytokines, you know, things also
like TNF-alpha, or…
[Matt]: IL-6. IL-1 even more so, but IL-6
is part of that equation, too. We’ve known
this for some time that if you infect an animal,
it will create an immune antibody response,
a cytokine cascade. Those cytokines have a
direct communication pathway into brain structures,
including things like the hypothalamus, which
regulates sleep. And it’s that immune cascade
that is actually a trigger in dialing up the
amount of sleep. So when animals get sick,
just like when humans get sick, the thing
that you want to do is just curl up in bed
and go to sleep.
And in fact, you get sleepier as a consequence
of being infected. Why is that? The reason
is because there is nothing better…Sleep
is the Swiss Army knife of health, that no
matter what the ailment…
[Rhonda]: That’s so cool. I don’t know…
[Matt]: You know, there is something more
than likely in the armament of sleep toolbox,
sort of as it were, that will deal with that,
and so, and I promise I’m not going off an
attention here, so the idea is that these
immune factors are sleep-instigating.
[Rhonda]: This is awesome. Wow.
[Matt]: They are sleep, sort of, you know,
they’re somnogenic, you know. Somno-sleep-genic
increased genesis improvement. So what I’ve
been thinking about is whether or not the
sauna, the benefit of the sauna, is both thermal,
but is also, by way of this powerful immune
[Rhonda]: It does.
[Matt]: …that you get this.
[Rhonda]: That’s phenomenal.
[Matt]: And I’ve read some of the studies,
and I’m blanking on his name, you will probably
remember his name, they’ve looked at some
of essentially sort of, you know, the sauna-induced
[Rhonda]: Yeah, so that Charles Raison has
been doing that work.
[Matt]: Thank you. Yeah.
[Matt]: Thank you. Brilliant. Yeah, so I’ve
read some of Charles’s papers.
[Rhonda]: Yeah, we had him on the podcast.
He was talking about how the sauna induces
IL-6 and some of that stuff. And the same
thing that happens with exercise. So that’s…
[Matt]: Yes. So that’s why I mentioned exercise,
which is that I think…
[Rhonda]: Wow, that’s so fascinating.
[Matt]: So now we’re all starting to sort
of, you know, realize how the sleep system
is augmented, one of those paths, you know,
is light. The other is darkness. The other
is temperature. But here’s a fourth one. The
immune system. And we know that you can inject,
you know, some of these cytokines into animals,
and you can almost induce sleep. It’s that
[Rhonda]: That’s so cool.
[Matt]: So now I’m starting to think, I wonder
if some of the sauna-based benefits and some
of the exercise benefits, because when you
exercise, you also typically get some of these
pro-inflammatory cytokines that sort of get
released to perhaps deal with some of, you
know, the essential distress. And, you know,
I think we’ve both spoken about this that…
[Rhonda]: So that way, like if you’re exercising
a lot, you do seem to require more sleep,
or you sleep more?
[Matt]: Well, we don’t know that, but that’s
my current theory…
[Rhonda]: Yeah, it’s a great hypothesis.
[Matt]: …which is that, you know, you possibly…
[Rhonda]: Are you gonna test that? That’s
[Matt]: Yeah, we’re going to test. So we’ve
just actually been looking at studying not
with exercise or with saunas but we’ve had
been looking at sleep and pain. And when you
deprive people of sleep, you get a chronic
release of these pro-inflammatory cytokines,
which is not a good situation. Acute, great,
somnogenic, good for the body. But chronic
long-term, we’re starting to piece together
a brain-body-sleep-pain interaction, which
I think has, we haven’t published this yet,
but should have marked implications for the
hospital environment. Because the one place
where you do not get a good night of sleep,
where it’s architected against the night of
sleep, is the one place where you need sleep
the most, and it’s usually the one place where
you are in pain the most.
And I want a revolution to happen regarding
sleep in hospitals. And I’m desperately trying
to work with folks. For example, the NHS in
the United Kingdom, National Healthcare System
there. I would love to work with any hospital
system in the U.S. to solve this crisis, I
think, of sleep in the hospital environment.
But, to come back to your point, this is a
great experiment to do. You know, is exercise-induced
acute inflammation, and is sort of, you know,
heat/shock sensitive, you know, induction
of immune responses by way of saunas or hot
baths, are those a pathway triggering good
sleep? And if so, can we find novel therapeutics
for, you know…
[Rhonda]: I’ll put you in touch with Dr. Jari
Laukkanen too. He’s in Finland.
[Rhonda]: He’s, like, the leading researcher
on saunas and…
[Matt]: Has he done some of the stuff on longevity?
[Rhonda]: He’s the guy doing all the longevity,
Alzheimer’s disease, cardiovascular disease…
[Matt]: Yes, yeah, yeah. I’ve read some of
his work. Yeah, yeah, yeah.
[Rhonda]: Yeah, I mean, and he’s a friend
of mine as well. So I’ll put you in touch
[Rhonda]: Because that would be really cool
if you guys could…
[Matt]: If we can start tracking their sleep,
you know, and see perhaps. And you can sort
of put this into a statistical sort of triangulation,
which is called the mediation analysis. And
you could see, you know, what’s causing that
longevity benefit. Because we know for example
that the shorter your sleep the shorter your
life, that short sleep predicts all-cause
mortality. It’s also probably one of the most
significant lifestyle factors determining
whether or not you’ll develop Alzheimer’s
disease. And all of these things I know have
been linked to, for example, sauna, which
is longevity, decreased susceptibility to
development of dementia and cognitive decline,
is part of that…I’m not trying to say it’s
all about sleep. But, the part of it about
sleep, you know…
[Rhonda]: Really, you know, one of my major
interest is aging, increasing health span,
as well as performance, without the trade
off, and every time I always come back to
sleep. The studies, I mean, it just constantly
coming back in my face how important sleep
is for all those things.
[Matt]: It’s the superordinate lever that
transacts all health benefits, you know. It’s
the tide that rises all of the health boats.
And I think it’s wonderful that folks, you
know, here in Silicon Valley are going after,
you know, individual pathways of disease,
trying to manipulate, you know, immune pathways
or trying to manipulate mTOR for sort of,
you know, longevity.
But I think what’s been missing in this equation
is that there is, you know, an Archimedes
lever here. There is one superordinate node,
that if you pull that lever, all of the other
pathways are instigated, activated. And that
superordinate node or lever is this thing
called sleep. You know, there is no physiological
system that we’ve been able to measure that
isn’t wonderfully enhanced by sleep when you
get it, or demonstrably impaired when you
don’t get enough.
[Rhonda]: Right. I kind of wanted to go back,
because you were talking about the effects
on how the immune system can be, you know…what
did you call it? I mean, it induces sleep.
[Rhonda]: Somnogenic, yes. How it’s somnogenic.
So interesting. I had no idea that the cytokines
were responsible for that, and it makes perfect
sense. So, anyway. But also, so the opposite
is true where sleep also affects the immune
system, or a lack of sleep also, right?
[Matt]: It does, yeah.
[Rhonda]: And that is something, you know…Your
immune system is the first line of defense
against, even pretty much everything, right:
cancer, viruses, bacteria.
[Rhonda]: And I think in your book, you were
talking about the effects on, some studies,
on natural killer T cells. I mean, I was just,
like, blown away by some of the numbers.
[Matt]: This is a frightening study. So it’s
done by my colleague Mike Erwin. You take
a group of individuals, and you’re not going
to deprive them of sleep for an entire night.
You’re simply going to limit them, restrict
them to four hours of sleep for one single
night. And then we’re going to measure the
amount of reduction in natural killer cell
So just to take a step back, natural killer
cells are a critical part of your immune defense
arsenal. And today, both you and I and everyone
listening to this podcast, we all have cancer
cells that have emerged in our bodies. But
typically, what prevents those cells from
becoming this disease that we call cancer
is, in part, these natural killer cells. So
what you wish for is a virile set of these
sort of immune assassins, these sort of James-Bond-like,
you know, that they will annihilate these
foreign organisms. You want a virile of those
at all times.
So take a group of healthy people, limit them
to four hours of sleep for one night, and
what you see is a 70% reduction in natural
killer cell activity, seven zero. That is
an alarming state of immune-deficiency. And
it happens quickly essentially after one bad
night of sleep. So you can imagine, you know,
the state of your immune system after weeks,
if not years, of insufficient sleep. And it’s
now the reason I think that we probably are
finding, at the epidemiological level, significant
links between short sleep duration, not getting
enough sleep, defined as six hours or less,
and your risk for the development of numerous
forms of cancer. Currently, that list includes
cancer of the bowel, cancer of the prostate,
cancer of the breast.
And the link between the lack of sleep and
cancer is now so strong that, recently, the
World Health Organization decided to classify
any form of nighttime shift work as a probable
carcinogen. So in other words jobs, that could
induce cancer because of a disruption of your
So, you know, that’s the immune system in
cancer. But it doesn’t stop there as well.
Another great study that was done by Aric
Prather [SP], who is over at UCSF, another
good colleague of mine, he did this brilliant
study that I write about in the book. He basically
measured the sleep of a group of healthy people
for a week before using these wristwatches
that are accurate. And then he quarantined
them in a hotel, in a set of hotel rooms.
And then he proceeded to stuff up their nose,
squirt up their nose, rhinovirus, essentially,
a flu virus. And then he quarantined them
for a week. And he measured how many of them
became infected. And he was measuring all
sorts of stuff. He collected every ounce of
snot that they blew out of their nose, all
of the mucus, everything.
And what he found was that those people who
were getting five hours of sleep or less in
the week before they came in and were infected
relative to those who are getting seven hours
of sleep or more, those people who were getting
five hours of sleep in the week before they
got infected were four times more likely to
end up developing the flu than those people
who are getting seven hours or more.
The final nail in that sort of immune coffin
for me is a study that was done by Eve Van
Cauter, who’s a wonderful endocrinologist.
And I’d love, by the way, to speak about sleeping
diabetes and glucose regulation, too. She’s
doing some great work there. But she did a
great study. She looked at the amount of sleep
that you were getting in the week before you
get your flu shot.
And what she found is that if you’re getting
sort of less than five or six hours of sleep
in the week before you get your flu shot,
you only produce half of the normal antibody
response, rendering that flu shot largely
useless, which stuns me that, you know…And
here’s his work technology could revolutionize
healthcare. You know, what if Kaiser, you
know, had access to a sleep tracker or had
its own sleep tracker. And Kaiser, for those
not familiar, is a health care provider here
in the United States.
But what if, you know, any healthcare provider
or system in any country had access to your
sleep in a non-Big-Brother way. And it was
tracking your sleep. And through an app, it
would say, “Hey, you’ve had a great week of
sleep this weekend in December,” or “this
week in November. Now is the time to come
and get your flu shot. I’ve listed out three
appointments. Just tap on the one that you
Or it says, “Look, I know that you scheduled
yourself for a flu shot. You haven’t quite
got the good sleep that you need this past
week. Let’s try again next week.” And if you
get the sleep next week, then you give them
access to getting the flu shot. Because otherwise,
we’re wasting money. And the flu, you know,
cost the United States, the flu season cost
the United States about $10 billion directly
in terms of healthcare burden. Even if I could
nudge that by just 1% or 2% by understanding
people’s sleep and helping architect a system
that co-opts around sleep time and management,
we could save hundreds of millions of dollars
to the U.S. economy.
[Rhonda]: Wow. Speaking of the flu, I think
I even told you this in a previous conversation,
like, I almost never get sick. Maybe once
a year, I’ll get a little runny nose for a
couple of days. I’ve got a really good immune
system generally speaking. But this last year,
I had a baby, and along with having a baby
comes sleep deprivation. And there’s absolutely
nothing I can do about it. I mean, you have
to wake up multiple times to feed your baby.
And so that’s what I was doing.
And, I mean, I got the flu. I got a cold.
I mean, I was sick, like, every month. I mean,
it was crazy. I’ve never experienced…of
course, my baby luckily was getting all my
antibodies and didn’t get sick, thankfully.
But it was, like, completely connected to
[Matt]: It’s impressive, isn’t’?
[Rhonda]: Yeah. I mean…
[Matt]: It’s the same wisdom that, you know,
mothers would sort of say, you know, that
speech. You’re getting sick a lot. Are you
sleeping enough? You know, and I’ve often
thought that people like me, sleep scientists,
all I’m really doing is putting the data behind
everything that your mother and Shakespeare
ever told you about sleep. You know, I’m just
putting data. They have knew it all long before
the data came along. But you see that exact
same thing in parents. Sickness is rife in
those early years. You see it in medical resident.
I think a friend of yours, who I enjoy this,
too, a lot, Peter Attia, was saying this about
when he was in medical residency, and you
see this in residence, too, you know, sickness
rates, you know, increased dramatically. There
is weight gain, the diabetic profile, you
know, in terms of their ability to manage
glucose becomes markedly impaired. We know
that testosterone in men plummets when sleep
gets short. In fact, men, who are sleeping
five hours or less will have a level of testosterone
which is that of someone 10 years their senior.
[Matt]: Which, in other words, a lack of sleep
will age men by a decade in terms of that
aspect of virility and wellness.
[Rhonda]: How quickly does that happen? Do
you know? Is that like…?
[Matt]: So you can see that within almost
days once you start to dose people on that.
It comes about quite rapidly. You can see
hormonal change. I mean, I think that’s what…if
there is a major access…oh, sorry, axis
within the body that is altered by sleep it
is the hormonal axis. I think that principally,
through the root of the autonomic nervous
system, that is the lens through which almost
all sleep and sleep deprivation effects can
And I’m just starting to write up a theory
paper about this, which is an encompassing
theory about sleep deprivation within the
human brain and the body. And I think there
is a common, central dictating governing pathway,
which is the autonomic nervous system through
which there is a manipulation of many of the
hormone systems, which then give rise to the
whole host of deficits that we see by way
of a lack of sleep, which then ultimately,
through chronic exposure of sleep loss, give
rise to all of the diseases that we know are
associated with a lack of sleep. And every
single disease that is killing us in the developed
world has causal and significant links to
a lack of sleep.
[Rhonda]: One that I’m particularly interested
in is Alzheimer’s disease. It’s something
that I’ve been researching for a while. I’m
about to get published thankfully soon.
[Rhonda]: Thank you.
[Matt]: Was this a review paper?
[Matt]: Oh, please send it to me.
[Rhonda]: And it happens to do with a gene
called apoe4. And I found out that I have
one copy of this of this allele. And when
I found that out years ago, I was like…it’s
probably the biggest risk factor for late-onset
on Alzheimer’s disease besides age.
[Matt]: Yes. So I think one of those alleles,
it’s about sorts of two to threefold risk.
[Rhonda]: Two to threefold.
[Matt]: If you have two of them, it’s like
an 8 to 12.
[Rhonda]: Yeah, it’s, like, pretty bad. And
25% of the population in the United States
has at least one allele. So it’s definitely…And
just because you have it doesn’t mean you’re
necessarily gonna get Alzheimer’s disease.
[Matt]: That’s right.
[Rhonda]: Not everyone has it. But there’s
a huge interaction with, you know, diet and
lifestyle. Probably the biggest lifestyle
interaction with this gene is sleep. And so
that was that was where I became very interested
in how sleep affects the brain, and how it
affects, you know, Alzheimer’s disease, and
all that. So I’d love to kind of talk a little
bit about that.
[Matt]: Yeah, so we’ve been doing a lot of
this work. We have a large research program
here at UC Berkeley at the Sleep Center that
is devoted to aging and Alzheimer’s disease.
And we’ve been very fortunate to get many
grants from the NIH here to study this. I
think that the story is fascinating because
it’s a bi-directional relationship between
sleep. And the pathology that we know is associated
with Alzheimer’s disease.
So in Alzheimer’s disease, there are at least
two protein culprits that we believe are underlying
of the brain pathology that seems to create
this thing called Alzheimer’s. One of them
is a sticky toxic protein called beta amyloid
that accumulates in these clumps outside of
brain cells. And that creates these amyloid
plaques that seem to be correlated with your
disease risk and disease severity.
The other is a protein that we know probably
less about, which is a protein called tau
protein. And that sits inside of cells. And
it creates a support structure for communicating
and funneling many of the critical ingredients
up and down your nerve cells to keep them
in rude health. And during Alzheimer’s disease,
that protein starts to sort of fall apart
and dismantle. And you get these sort of tau
tangles. And the structure of the nerve cell
and its ability to transport all of the ingredients
that it needs to operate starts to collapse
and fails like a tunnel collapsing down.
So one of the discoveries that we made back
in 2013 was that I was looking at the distribution
of this sticky toxic protein called amyloid
in the human brain. And what’s fascinating
is that it doesn’t build up in the brain in
homogeneously. Amyloid builds up in some parts
of the brain far more severely and early in
the course of Alzheimer’s disease. And other
parts actually remain completely uninvaded
by this thing called amyloid. In fact, parts
of the motor cortex, for example, or parts
of the visual cortex, you see almost no amyloid
in our Alzheimer’s patients. And that’s probably
the reason why their motor functions and their
vision is unchanged.
But one of the earliest places were beta amyloid
builds up and then built up most severely
in late stages is back, again, in that medial
prefrontal cortex that sits right there in
between the eyes. Why was I interested in
that? I was interested because when we were
doing studies where we would map, with all
of these electrodes over your head, we would
map the deep sleep that you were having.
And we could do some clever mathematical modeling
of those deep sleep brain waves. And we could
try to triangulate where was the electrical
epicenter of those electrical deep brain waves
of deep sleep. And it seems as though they
come from all over the brain, but the principal
epicenter that generates your deep sleep sits
right there in the middle part of the prefrontal
cortex. It is exactly the same part of the
brain that accumulates toxic beta amyloid
Then we’ve done studies, and other people
have done studies before us, that demonstrated,
as we age, our sleep gets worse. But just
any type of sleep. Especially that deep quality
of sleep that we know and we spoke about is
critical for saving and learning and retaining
new memories. So all of these jigsaw pieces
started to get put together in my head. I
thought we need to do some studies. Is it
possible that the amount of amyloid that you
have in the brain in this sleep-generating
center, it should directly predict the deficit
in the amount of deep sleep that you get?
If it predicts the deficit in the amount of
deep sleep, it should predict the deficit
in your ability to hold on and retain new
memories, which is a hallmark cognitive feature
of Alzheimer’s disease, difficulty learning,
So we did the study, and it’s exactly what
we found. The more beta amyloid that builds
up in this central frontal part of the brain,
the less the deep sleep that you have. The
less amyloid-related deep sleep that you had,
the more forgetful you were the next day rather
than the more that you remembered. So this
was the first part of the Alzheimer’s sleep
equation, which is that Alzheimer’s disease
attacks the deep-sleep-generating regions
and you have a diminution of deep sleep, which,
in turn, blunts your learning and memory abilities
and you become more forgetful.
A far more important discovery was made by
an another group, far more important than
the one we made, which was essentially the
reverse direction, which was to say, “Rather
than amyloid sort of decreasing sleep, could
sleep actually decrease the amount of amyloid
that you get?”
And this is a discovery that was made in rats
back in 2009, I believe, was the first evidence
that was published in science. A colleague,
[inaudible 01:24:09], who is out on the east
coast at the University of Rochester, and
she made two wonderful discoveries. The first
was that we’ve known for a long time the body
has a waste sewage system called the lymphatic
system. But the brain doesn’t have its own
lymphatic system. The lymphatic system does
not penetrate the brain. So where does all
of the garbage, the metabolic garbage go that
your brain cells produce? Where is the sewage
system for the brain?
And she discovered it. It’s actually made
up of a set of cells called glial cells, which
are these supporting brain cells. And so she
called it the glymphatic system rather than
the lymphatic system. So your brain does have
its sewage system, this glymphatic system,
and that’s the discovery that she made. Remarkable.
Then, and I’m not quite sure what motivated
her to do this, she started to measure how
efficient that glymphatic, that waste system
was when the rats were awake and when the
rats were asleep. And what she found was that
it’s during deep sleep that these brain cells
actually shrink by almost 60% when we sleep.
Blows my mind.
[Matt]: It’s almost like, you know, all of
the buildings in New York all of a sudden
shrink, and it leaves these much greater,
large areas for the cleaning crews to come
in and clean up all of the metabolic detritus
of the city’s activity during the day. It’s
exactly what happens during sleep. And the
cleaning solution is what we call cerebrospinal
fluid. And through a pulsatile mechanism during
sleep, you get a 10% to 20% increase in the
bathing of cerebrospinal fluid through the
brain, which washes away all of the metabolic
byproducts that have been building up. One
of those metabolic byproducts is beta amyloid.
And, in fact, if you deprive those rats of
that deep sleep, you immediately get an increase
in toxic beta amyloid.
So now we’ve linked these two. I’m sorry it’s
a long story. But if you’re not getting
Enough deep sleep at night, you’re not giving
yourself the chance for the kind of good night
and sleep clean process to remove the beta
amyloid. So more beta amyloid builds up. Where
does it build up? Tragically, in the very
same regions of the brain that generate the
deep sleep that you need to clear out the
toxic amyloid. So you start getting less deep
sleep, so you get more toxic protein, more
toxic protein, less deep sleep, less deep
sleep. It’s a self-fulfilling prophecy, and
it’s a nonlinear exponential curve.
If you look at how amyloid builds up in the
brain, and if you look at the trajectory of
Alzheimer’s disease, it is a nonlinear exponential
curve. It fits exactly what the sleep-dependent
model of amyloid clearance would predict.
If you’re not getting sufficient sleep, that’s
the reason why now insufficient sleep seems
to be one of the most significant lifestyle
factors determining that.
Now you could say, by the way, those studies
were in rats, and you deprived from the sleep
for one night. What about humans? Like, surely…Well,
the study has now been done. Great study done
out of Wash U by a team of scientists. They’re
led by David Holtzman. And they took a group
of humans, and they did this very clever method
where they deprive them of deep sleep but
they didn’t deprive them of sleep.
And you think, it sounds paradoxical. I can
play you these auditory tones. Now this is
not like the memory reactivation where you
play a tone and then you leave the brain alone
for a while. Here, I’m just going to keep
playing tones to your brain, really sort of
annoying tones. But I can play them at a level
that doesn’t wake you up, but it lifts you
out of deep sleep and keeps you in shallow
So what’s delightful about this method is
that I can selectively excise one type of
sleep, deep sleep, but I don’t wake you up.
So there’s no stress of awakening. You are
asleep for the same amount of time, but the
quality of sleep is decreased.
[Rhonda]: Can street noise do that?
[Matt]: We don’t know. Although I will come
back to that when we speak about hopefully
sleep appetite regulation, sleep glucose regulation,
and sleep in low socio-economic cultures…
[Rhonda]: This is a little terrifying.
[Matt]: And I think that it’s possible. I
think there’s other factors that link poor
sleep in low SES, socioeconomic backgrounds.
Is noise pollution one of them? I actually
think it is. Untested as yet. But what they
did with these human participants, they selectively
remove deep sleep while keeping them asleep
so total sleep time has not changed. And then,
in the morning, they woke them up, they rolled
them over, and they did a spinal cord puncture,
a lumbar puncture, and they measured the cerebrospinal
fluid that was populated within the spinal
cord, which also goes around the brain. And
you can measure the amount of beta amyloid,
which is a reflection of perhaps how much
amyloid is there within the brain.
After one night of essentially a loss of deep
sleep, you saw an immediate rise in the amount
of beta amyloid. So it is a causal manipulation.
That insufficient sleep in rodents and in
humans will lead to a rise in beta amyloid.
[Rhonda]: Yeah. I think it was like 25% to
[Matt]: It was.
[Rhonda]: Yeah, it was definitely…So, of
course, that’s all I could think about the
past nine months…
[Matt]: Showing that, yeah.
[Rhonda]: And the buildup of amyloid.
[Matt]: You know, I would tell you, that’s
the last thing you should think about. You
know, try not to worry too much about it,
you know, because it’s not to say to diminish
the loss of sleep but simply to say that worrying
about it is not gonna do you any good.
[Rhonda]: Of course.
[Matt]: But how do you tell that especially
to people who, you know, have one ApoE allele…apoe-4,
sorry, allele. What I would also say that’s
important for people, if you know your ApoE
status, and if you are apoe-4, be mindful
of snoring as well. Because people who are
apoe-4-positive, they also have a significantly
elevated risk of a sleep disorder that we
call sleep apnea, which is sleep disorder
breathing, which is heavy snoring and a cessation
of breathing entirely. And then you gasp when
you wake up again.
Heavy snoring sleep apnea is a killer. It
is an outright killer. It increases your risk
of basically everything you don’t want: cardiovascular
disease, stroke, diabetes, and obesity. It
also increases your risk of immediate death
through a higher risk of car accidents.
But one of the other problems with sleep apnea
is that you don’t get the amount of deep sleep
that you need. And you have hypoxic damage.
Because you stopped breathing, your oxygen
saturation goes down. You get hypoxia damage
particularly in a region that is most sensitive
to it in the brain, which is, drumroll, the
hippocampus, the very same memory structure
that is attacked in Alzheimer’s disease.
So now you can see why I appeal for this sensitivity
in this danger to sleep apnea. Because if
you are apoe-4, you’re already at high risk
of Alzheimer’s disease, you need to pay attention
to your sleep. If you start snoring, and you
have sleep apnea untreated, you will get less
deep sleep. So you’re compromising the thing
that you need to try and lower your amyloid
risk to begin with, because you’re going to
build up that amyloid, because you’re not
going to get the amyloid clearance elsewhere
in the body, for example, in the liver. And
then where still the part of the brain that
is attacked severely by Alzheimer’s disease
and atrophies, which is the hippocampus, which
is why memory fades, is a part of the brain
that is damaged when you stop breathing because
of oxygen desaturation.
So, as an appeal, even if you are not apoe-4-positive,
but you are snoring, or you know someone who
is snoring, go and see your doctor and get
a sleep apnea test. It is potentially life-saving.
[Rhonda]: Actually, I had a sleep apnea test
because of my night awakenings. I didn’t know.
It was, like, maybe I have apnea and I couldn’t…And
so I’m having this terror. And of course,
that wasn’t it. I don’t have sleep apnea.
But I didn’t know about the connection between
apoe-4 and sleep apnea. That’s terrible because…
[Matt]: It’s quite a dramatic…
[Matt]: I mean, you can bind those two together.
It’s a real…I mean, it’s gasoline on an
already started fire.
[Rhonda]: People with apoe-4 also don’t repair
damage and their brain as well as people with
[Matt]: They don’t.
[Rhonda]: So you’re talking about hypoxic
damage in addition, you know, there’s sort
of just this potentially compounding effect.
[Matt]: And I think that’s, you know, that’s
part of the reason…I mean, so I have some
family members that have been associated with
Alzheimer’s disease, one recently passed away.
And it’s been a big motivation for me personally.
I started, you know, 20 years ago. My PhD
was looking at people with degenerative dementia.
That’s how I actually got into sleep, because
I was seeing these sleep abnormalities.
And, you know, I’m so desperately trying to
find ways to help and to combat that incredible
epidemic of dementia. And that’s why we’ve
started to try and develop things like electrical
brain stimulation methods, to try and help
actually augment human sleep, an electrically-supercharged
sleep, as it were, to see if we could give
back some healthy quality of deep sleep to
patients who are aging or those with dementia.
Can we essentially, you know, amplify the
amount of deep sleep? And in doing so, can
we salvage aspects of learning and memory?
But better still, could we actually start
to increase, you know, the amount of glymphatic
clearance of beta amyloid?
Now I suspect that when the disease is in
play and you’ve been diagnosed in those late
stages, I don’t know how much sleep is electrically
charged even as it maybe could help. What
we’re now trying to do in our studies is actually
retrospectively find out, is there a particular
decade of life or decades of life when a decline
in sleep makes you most susceptible to then
developing a lot more amyloid later in life?
In other words, we’re trying to now identify
these vulnerability windows during the lifespan.
The reason I want to do that is, if I can
scientifically convince myself of a knowledge
base of vulnerability sensitivity to insufficient
sleep, my guess is that it’s just across the
lifespan, it gives me a chance to know where
is the inflection point of not late-stage
life treatment but early-life prevention.
Because that’s what medicine has to do right
now, I think. We’ve done a good job at extending
lifespan but a miserable job at extending
health span. Lifespan is probably about treatment.
Health span is probably about prevention.
Sleep needs to be part of that discussion.
And sleep is usually absent in many of these
conversations for either lifespan or health
span despite it having a demonstrable impact
on both. But my hope is that to be able to
find that sensitive time when your risk for
Alzheimer’s development by way of insufficient
sleep is present. That’s where I go in and
start augmenting your sleep with electrical
brain stimulation or other methods that we’re
trying to develop as well.
[Rhonda]: I think, at least for the apoe-4-positive
individuals, by the age of 40, the amyloid
plaques start to really…
[Matt]: That’s right, yup.
[Rhonda]: …you know. So, to me, it would
seem that, you know, before 40, and certainly
when you hit 40, you better have your sleep
[Matt]: I would say before. I mean, we can
see the decline of deep sleep occurring in
people in their 20s. That’s when your deep
sleep starts to decline.
[Matt]: Which is, you know, frightening, isn’t
it? It’s sad. And at that point, it’s in the
mail. You know, by the time you’re 50 years
old, you’ve lost about 50% of the deep sleep
that you are having when you are a young teenager.
By the time you’re 70%, there’s only about
5%…sorry, by the time you’re 70 years old,
there’s only about 5% of your deep sleep left
that you had when you were young and healthy.
By the time you’re 80, we almost can’t detect
any of these deep sleep brain waves anymore.
[Rhonda]: Do you know if there is changes
in the way your core body temperature regulates
with age as well?
[Matt]: There is. So what happens with age
is that your circadian rhythm that we’ve spoke
about before…I’m just going to…So your
circadian rhythm that we spoke about before,
which in healthy people is nice and high and
peaking during the day, lots of activity,
and then drops down at night, lots of inactivity,
lots of deep sleep, that sinusoidal wave starts
to flatten out as if someone has compressed
it as we get older. So our circadian rhythm
gets weaker. So we get we feel sleepier during
the day and not as alert as we used to do,
but we don’t feel as though we are as sleepy
at night. We’re more awake at night because
our circadian rhythm is blunted.
And so I think that’s another area for aging
intervention, is how can we modulate the circadian
rhythm. And it turns out that it comes back
to light. So one of my colleagues in the Netherlands,
Eus Van Someren, did a great study where he
installed circadian-regulating light in an
elderly care home, you know, in a home where
you care for the elderly. And many of them
had cognitive decline. And he’s now done some
of these studies with people in with Alzheimer’s
disease in some of these care homes.
And when you start to create appropriate lighting
in the internal environment in these homes,
which if you go into them, and I used to go
into them all the time during my PhD, these
sort of care homes, and I would be testing
my patience and seeing my patients there,
they were grim environments with dim light.
And these patients never got outside. They
rarely had a window. And Eus Van Someren,
the scientist in the Netherlands, he was able
to really produce this strong bout of light
during the day inside the care home and then
drop out that light in the evening. And he
regularized their light. And he improved the
circadian rhythm. And he improved cognitive
[Matt]: Cognition got better in these Alzheimer’s
[Rhonda]: That’s cool.
[Matt]: Now you can look at that same manipulation,
that same lighting hack, at the beginning
of life. So in the neonatal intensive care
unit, what we used to have when we go in there,
you just have constant light on all of the
time, dim light on 24 hours a day, which prevented
those infants getting the signal of a regulating
24-hour light-dark cycle.
If you regularized light in the neonatal intensive
care unit, so light during the day, darkness
at night, in these studies, what you see is
basically almost a 50% to 60% increase in
oxygen saturation within the blood of these
neonates in the intensive care unit. Weight
gain increases dramatically. And they end
up exiting the neonatal intensive care unit
at about five weeks earlier than they would
Circadian regulation of sleep leads to better
health outcomes. It does so in neonates. It
does so in the elderly. So I think there is
all manner of sleep possible interventions
that we can think about across the lifespan
when it comes to modifying disease risk at
any stage of life.
[Rhonda]: Yeah. And I think, at least as far
as I’ve, you know, understood from our conversation
thus far, some of the main things for prevention
in optimizing your sleep really are preventing
that emotional stimulation, that anxiety,
and getting particularly a few hours before
bed, the bright LED exposure early in the
morning, and then not having the bright light
exposure, you know, about four hours before
bedtime. And then the cold, like, having your
body temperature go down. So, you know, whether
that’s through something like, you know, having
a hot bath or a hot shower or even possibly
the chili pad, which I’m gonna actually experiment
[Rhonda]: Because now I’ve been using this
Oura ring that I’ve been tracking my sleep.
By the way, do you know how accurate, like…So
it tells me, in my app, the light sleep, the
REM sleep, and deep sleep. And I know there’s
nothing going in my brain measuring any sort
of brainwaves. So it must be movement, right,
[Matt]: Yeah, I won’t say specifically even
though I know the folks at Oura and I know
the folks at almost all of the sleep-tracking
companies, most of them right now use a combination
of your heart rate, which you can pick up
through the pulse waveform either on the wrist
or on the finger, or if it’s a bed sensor,
through the heartbeat, as well as respiration.
And they’ll use a combination of those things
together with movement to try and stage your
Right now, without naming, you know, I don’t
think there’s anyone that is necessarily better
than the other, at least none of them have
published scientific evidence. We sleep scientists
have actually looked at these devices relative
to gold standard, what we call polysomnography,
which is essentially like doing a sleep study.
If you were to, here at my sleep center, we’ve
got all of this equipment. You look like a
spaghetti monster with electrodes on your
[Rhonda]: I don’t know who can sleep with
[Matt]: Yeah, I know. But we can measure that
sleep with high precision, high fidelity.
That’s the gold standard. And when you compare
these sleep trackers, unfortunately, they’re
not quite accurate. They either overestimate
or underestimate sleep onset latency, how
long it takes you to fall asleep. They either
overestimate or underestimate the amount of
time that you’ve been awake. Sleep duration
may not be bad with some of them, total sleep
duration, but once you get into the sleep
stages, that’s where things become more inaccurate.
Their ability to separate non-REM from REM,
it is getting better. It could be in the region
of sort of 60% accuracy, possibly lower. But
right now, we’re not there yet. Will we be
there in about four years’ time? I actually
think we will be.
[Rhonda]: Because that’s what…I’m really
interested in the deep. Obviously…
[Matt]: But also I would say REM sleep, too,
you know, I think REM sleep takes a backseat.
At sort of, you know, a bit of a neglected
stepsister in the sort of sleep conversation.
REM sleep, we found all manner of different
functions, one of which is emotional first-aid.
It’s incredible for palliative emotional benefit.
It’s not time that heals all wounds, but it’s
time during REM sleep that provides emotional
[Rhonda]: I guess that would also then be
important to help you sleep better, because
if you are managing your emotions better,
you probably have less anxiety…
[Matt]: Correct. Exactly. But also REM is
for the body, too, you know. It seems to regulate
cardiovascular function. So REM sleep is critical
[Rhonda]: Oh, really? Is REM what’s important
for lowering the blood pressure?
[Matt]: No. Well, it’s during deep sleep that
you get this lowering of blood pressure, and
we’ve got some data right now that we’re about
to publish that deep sleep provides actually
a homeostatic recalibration of blood pressure.
What do I mean by that? If I measure your
blood pressure before sleep and then after
sleep at a matched circadian time, so we remove
the circadian fluctuation of which there is
a large one to your systolic blood pressure,
but we match it, and so the only thing that’s
different is the quality of your sleep, what
we’ve discovered firstly is that the amount
of deep sleep that you have measured in these
big delta slow waves during deep non-REM sleep,
that predicts how much of a drop in your systolic
blood pressure you will have in the morning
relative to the evening as if deep sleep provides
a recalibration of the cardiovascular system.
[Matt]: If you’re having higher frequency
brainwaves that are not like the deep sleep,
that’s what we call unrestorative sleep, if
you have a ratio of very little deep brainwave
activity and a lot of high-frequency wake-like
brain activity, what we call delta-beta ratio
that predicts very bad cardiovascular outcomes,
that if anything, your systolic blood pressure
is even worse after sleep than it is the night
And what we found is that during aging, now
we looked at this and healthy people, then
we replicated in older adults, and we found
that the older that you get, the less deep
sleep that you get, the more sort of wake-like
or faster-frequency brainwave activity, which
isn’t good during deep sleep. And that accurately
predicts the cardiovascular dysfunction that
we see in aging. So we now think we understand
in part why poor sleep is linked into poor
One of the benefits of sleep is that it resets
cardiovascular tone and particularly systolic
tone. It doesn’t stop there. There’s lots
of other benefits, heart rate drops, etc.
But REM sleep also seems to have this strange
function where you go through these cycles
where you get a massive activation of the
cardiovascular system, and then it falls silent,
and then a massive escalation again. And you
get this really remarkable increase in heart
And heart rate variability has been used as
a metric of health outcome, that if you have
high current rate variability, it’s very good.
It predicts lots of health outcomes. That’s
what you get during REM sleep. So we should
be careful not to think of everything being
all about deep sleep. It’s about all stages.
Every stage of sleep that we have, we have
ascribed a function to. Which makes sense,
you know, during sleep…Sleep is the most
idiotic of all things in terms of a creation
from Mother Nature. You know, you’re not finding
a mate. You’re not reproducing. You’re not
eating. You’re not caring for your young.
And worse of all, you’re vulnerable to predation.
So on any one of those grounds, sleep should
have been strongly selected against in the
course of evolution.
The exact opposite is true. Every species
that we’ve studied to date sleeps. What that
means is that sleepers fought its way through
heroically every step along the evolutionary
pathway, which must mean that if sleep doesn’t
serve an absolutely vital function, it is
the biggest mistake that the evolutionary
process has ever made. And if any one of those
stages of sleep could be removed, because
it was not important, because I want to, you
know, emphasize this type of sleep because
that other sleep must not be important, I
promise you, mother nature would have thought,
I’m gonna excise that stage of sleep out because
you’re just too vulnerable.
But I think, in your position, of course I
would be concerned about my deep sleep, thinking
about things like apoe-4 Alzheimer’s risk,
too. But I think the bottom line is that no
matter which way you slice the sleep pie,
you just can’t shortchange sleep.
[Rhonda]: I just want to figure out, it’d
be nice to know what environmental factors
are affecting, you know, different stages
of sleep, you know. So it’s like, you know,
if I do X, then that’s gonna, you know, affect
my deep sleep or it’s gonna affect my REM
or it’s gonna make me miss, you know.
For example, if I were to go to bed, let’s
say I usually go to bed at 9 or 9:30, and
if I were to go to bed instead at 11:30 or
midnight, because I had a social event or
something, and my circadian rhythm, usually,
well, my circadian rhythm these days is my
son. Wakes me up at 6 a.m. But let’s just
say I could sleep longer if I…let’s say
I wake up at 6 a.m., but then I’m like, “Oh,
I went to bed at midnight instead of 9:30.
I’ll try to go back to sleep.” Would that
be okay? Would I be able to, like, recover
something that I’m missing from the 9:30 to
midnight? Do you see what I’m saying? Like…
[Matt]: Yeah. So…
[Rhonda]: Should I go back to sleep? Should
I try to go back or…?
[Matt]: You should try to go back to sleep.
So you should try to get the sleep that you
[Rhonda]: The duration, yeah.
[Matt]: Yeah, the duration. So there are essentially
four ingredients, and this is probably going
to be my next book, which is essentially the
four pillars of sleep: depth, duration, continuity,
regularity. If you shortchange sleep on any
one of those, you get a compromised deficit
in brain and body. So depth is what we were
speaking about. It really should be depth/quality
of sleep. You need both the depth of those
deep sleep brainwaves but you also need all
of these different brainwave oscillations,
these things called sleep spindles, and slow
waves, and the coordination between those
two. It’s all about the electrical quality
of your sleep.
So you can have the duration of sleep, you
can have eight hours of sleep, but if it’s
not of the right electrical quality or depth,
you get deficits. But you can get lots of
high quality sleep, but if you’re only getting
four hours of it, it’s not going to be enough.
So it’s duration. Depth, duration.
Then it’s also about continuity. This has
been probably, in the past five years, one
of the explosions in the sleep field. If you
were to get eight hours of sleep but across
a nine-hour period because you are awake for,
you know, 5 minutes here, 10 minutes here,
30 minutes here, 15 minutes here, that’s very
fragmented sleep, which, by the way, alcohol
is another thing that will fragment your sleep
very much like that. So the continuity of
your sleep is poor. It’s not continuous. It’s
And even if you get eight hours of sleep in
a nine-hour period but it’s fragmented versus
you get eight hours of sleep all in a nice
one bout within eight hours, so in both of
those scenarios, it’s the same duration of
sleep, maybe it’s even the same electrical
quality of sleep, but if it’s fragmented and
littered and punctured with many awakenings,
the continuity of sleep is poor. And that’s
not good either.
And then, finally, the part is regularity.
This means going to bed and waking up at the
same time no matter whether it’s the weekend
or the weekday. Go to bed at the same time,
wake up at the same time. Those are the four
key pillars that we know of for sleep. And
you can hold any three constant and manipulate
one of them and you get a deficit.
Now we can’t quite weigh the complete, you
know, is it, you know, this one is worth,
you know, 40%. This one is worth sort of 20%.
We can’t yet do that. But it’s very clear
that those are the four pillars of good sleep.
So I mentioned that just coming back to the
quality of sleep, when you position of the
quantity of sleep on the clock face, on the
24-hour clock face, at an inappropriate time
according to your natural innate preference,
and I say that specifically rather than a
particular time, you will not get the same
quality of your sleep. So an extreme version
of this is a night shift worker, where they
may get eight hours of sleep in bed but they
are sleeping during the day, not during the
night. So it’s a complete reversal. The quality
of their sleep is not good. They don’t get
the same amount of deep sleep nor do they
get the same amount of REM sleep.
But let’s kind of walk that back a little
bit to your scenario, which is much more subtle.
Let’s say that my chronotype, and your chronotype
simply determines whether you’re a morning
person, an evening person, or somewhere in
the middle, about 30% of the population is
an extreme morning type or a morning type,
about sort of 40% percent is sort of neither
strongly morning or evening, and about the
remaining 30% is an evening type, we call
them owls and larks.
So if you’re a morning type, you like to go
to bed, let’s say, like, 9 p.m. And wake up
at 5 or 6. If you’re an evening type, you
may want to go to bed at 1 a.m. And wake up
at 9 a.m. The next morning. So when I’m saying
the optimal position of sleep, I’m saying
with the optimal for your chronotype, this
is your chronotype…And by the way, it’s
genetic. You don’t get to decide whether you’re
a morning type or an evening type. It’s hardwired
into your genes. We know the genes. It’s not
your fault. If you’re an evening type and
you’re listening to this, and society, which
is strongly architected against your chronotype,
we reward and we favor the morning types,
it is not your fault. It’s not a choice. And
there’s not too much you can do about it.
You can push and pull the system by about
30 minutes, 45 minutes, but not much more.
So coming back to it, when we position that
sleep on that eight-hour period, whatever
is optimal for you, you will get a nice distribution.
You’ll get all of the deep sleep that you
want and all of the REM sleep that you want.
And earlier in the conversation, I said that
we go through these 90-minute cycles, we,
human beings, it’s different for different
animals, you always go into deep non-REM sleep
first and then you always have REM sleep second.
And that repeats every 90 minutes throughout
What changes, however, is the ratio of non-REM
to REM within those 90-minute cycles as you
move across the night such that in the first
half of the night, the majority of those 90-minute
cycles are comprised of lots of deep non-REM
sleep and very little REM sleep. In the second
half of the night, that balance shifts, and
now you get much more REM sleep in the late
morning hours and very little deep sleep.
The reason I’m saying this is because let’s
take someone who goes to bed at a substandard
amount of time. Let’s say that they go to
bed at midnight, and they’re going to wake
up at 8. But instead they have to wake up,
because they’ve got an early morning meeting,
so they go to bed at midnight and they wake
up at 6. And my question is how much sleep
have they lost? And your response is, well,
they’ve lost two hours out of the eight hours,
which means they’ve lost 25% of their sleep.
And your answer would be right in a way, but
it would be wrong as well, which is that they’ve
lost 25% of their total sleep but they may
have lost almost 80% of their REM sleep because
that’s the REM rich phase of the night.
And it also works the other way around, too,
that if you go to bed too late, you will…So
what happens is, because of the circadian
rhythm, the brain has a different appetite
for different stages of sleep on the 24-hour
clock face. In the late evening and the early
morning hours, the brain has a dietary preference
for deep sleep. And it doesn’t very much have
an appetite taste for REM sleep. Through to
the second half of the night, that’s when
it gets its appetite for REM sleep.
So if you start sleeping at 4:00 in the morning
and you wake up it in the middle of the day,
your brain has lost its appetite for deep
sleep. And so it won’t get much. Now you will
probably get much more REM sleep as a consequence.
So you’ve got to be really careful. That why…
[Rhonda]: So it’s almost like your chronotype
may even dictate in a way how much deep sleep
you may get.
[Matt]: The chronotype will…If you sleep
naturally and you get the amount of sleep
that you want, your chronotype will probably
mean that…your circadian rhythm is actually
shifted as your chronotype, too. So you will
still get, as an owl, if you get your eight
hours when you want it, you will probably
get a similar sleep architecture as a lark,
who has gone to bed five or six hours earlier,
because your circadian rhythms are five to
six hours different.
But you’re right in the sense that owls typically
try to go to bed early. But because they are
designed not to fall asleep at that time,
they’re just gonna lie in bed. Many owls think
that they have insomnia. They don’t. They’re
just not going to bed at the right time. Because
they get into bed, and it’s like a teenager,
whose rhythm is also shifted late, you tell
them to go to bed because you’ve got to wake
up for early school start times, but there’s
nothing they can do because their circadian
rhythm has shifted forward in time. And it’s
the same for owls.
So what will happen is that they will probably
stay awake for a while, then they’ll get into
deep sleep. But then they have to wake up
at an earlier time, and they will lose a lot
of REM sleep. REM sleep, for emotional well-being,
what the owls typically experience: depression,
low mood, anxiety. So I think we’re really
starting to put the pieces together on that
[Rhonda]: And so for someone like me, I’m
somewhere in the middle, like, when I don’t…pre-baby.
Now I’m I guess you would call a lark because
I’m going to bed at, like, 9. My optimal time,
if I can go to bed when my son does, and it’s
like, you know, if I go to bed, like, at 8,
then I get the longer duration. That way,
if I’m a little more fragmented, then I, like,
at least can make up for it somewhat. But
I’m up at 6 a.m. Now if it were up to me,
I’d like to sleep…I usually would wake up,
like, 8 a.m.
[Matt]: Yes, 8 a.m.
[Rhonda]: Like, that’s my natural time.
[Matt]: And the reason that you can probably
go to sleep sort of, you know, 9 p.m. Or even
8 p.m. Now is because you’re chronically sleep-deprived.
And as a consequence, you’ve built up such
as sleep debt that, you know, there is a lingering
what we call sleep pressure in your system.
But, yeah, I think fighting your chronotype,
we found, it comes with deleterious health
consequences. Increased risk for poor cardio
metabolic outcomes. Things like, you know,
C-reactive protein is higher. If you look
at, you know, A1C in terms of sort of a raw
shot review of blood glucose, your blood sugar,
not good. If you look at your propensity for
being obese or being overweight, also not
great if you’re an owl and you’re not sleeping
according to your schedule.
[Rhonda]: Telomeres are shorter.
[Matt]: Telomeres are shorter as well.
[Rhonda]: So let’s talk about your, because
a lot of interesting researchers come out
of your lab on the blood glucose regulation
front and, well, I guess more on the eating
[Matt]: And appetite and eating.
[Rhonda]: …and appetite. Yeah.
[Matt]: I mean, the way I see it, it’s all
part of the energy intake expenditure system.
And sleep, you know, if you think about like
a weighing scale between sort of energy expenditure
and energy consumption, sleep, if you’re not
getting it, just annihilates that balance.
And we can speak about any one of those things.
But, you know, I think the blood glucose story
and sleep is very very well-worked out now.
It started with epidemiological studies, where
we started to see that people who were sleeping
less than seven hours were at significantly
higher likelihood of either being diabetic
or going on to develop diabetes. Many of them
were already, you know, what we call a pre-diabetic
state, or they had what we now call sort of
And then the question became, well, is that
associational or is it causal? So the next
studies that happened in this was work back
in…starts in the 1990s by Evan Carter at
the University of Chicago. Wonderful studies.
It took a group of healthy people, started
to limit them to different doses of sleep
for a week, you know, five hours of sleep,
six hours of sleep, four hours of sleep, and
what she showed was that, essentially, after
one week of short sleep, your blood sugar
levels are disrupted so significantly that
your doctor would classify you at that point
as being pre-diabetic after one week of short
And the way that they do this is what’s called
a glucose tolerance test, where you are fasted,
and then you are given this sickly-sweet drink
of glucose, and then they are measuring from
your blood in the next three or four hours
how quickly is your body able to dispose of
that blood glucose.
So what happens when you drink or when you
eat a meal is that your blood sugar spikes.
And you don’t want that spike to stick around
very long. If your body is healthy, it deals
with that raised level of blood glucose very
quickly, and it brings it back down very quickly.
That’s a healthy profile. That’s what we call
good glucose management. So how good is your
body at disposing essentially of that glucose?
And the way it disposes it is that cells in
the body, including muscle cells, will suck
up that glucose. And it’s called your disposal
index, or your disposition index, it turns
So what she found, Evan Carter, with her studies,
was that, firstly, the way that your body
knows how to absorb glucose and suck in that
glucose is that there is another chemical
called insulin, which is released by the pancreas,
beta cells of the pancreas, and that insulin
will instruct the cells of the body to open
up special glucose channels to absorb the
glucose and your blood sugar drops, which
is good and healthy.
Firstly, what she found was that when you
are not getting sufficient sleep, the beta
cells in your pancreas stop being sensitive
to the signal of high glucose. So the beta
cells, which normally are listening for this
spike in glucose, and as soon as they sort
of, you know, hear…no, they’re not hearing
it. They’re sensing it. But as soon as they
sense the spike in glucose, they release insulin.
And that insulin will drop your blood glucose.
But those cells had become insensitive to
glucose, what we call sort of glucose insensitivity.
And so the beta cells of the pancreas stopped
releasing as much insulin. It didn’t release
enough insulin to drop blood glucose, so blood
glucose remained high. If that wasn’t bad
enough, we’ve since gone on to demonstrate,
and you can do this with really clever studies,
taking tissue biopsies, the cells of the body
including muscle cells and fat cells, their
receptors stopped being as sensitive to insulin.
So, firstly, you’re releasing less insulin
when you’re sleep-deprived. But what little
insulin you do release is not instructing
those cells to open up the channels to take
away the monsoon of the glucose that’s flowing
in the channels of the body. So on both sides
of the glucose regulation, on the release
of insulin, to instruct cells to absorb glucose,
and on those cells themselves to be sort of
instructed by insulin, those cells became
less sensitive to the insulin signal. And
so, as a consequence, your overall ability
to deal with glucose became far more degraded,
and blood glucose remained higher, which sets
you on a profile of looking pre-diabetic.
[Rhonda]: Couple that with the standard American
diet, and, you know…
[Matt]: And you’re off to the races in terms
[Rhonda]: Yeah, and eating late at night.
So, actually, are you familiar with the studies,
I know Dr. Satchin Panda actually is the one
who told me about this, melatonin is actually
what is responsible for shutting down the
pancreatic beta cells from producing insulin.
So I wonder how much of the sleep deprivation
the melatonin system is involved in that.
[Matt]: Yeah. So in those studies, what was
good is that they held constant circadian
rhythms, they held constant ambient light,
and they even held constant physical activity.
The control group was sort of they were in
bed for eight hours, but the people who are
sleep deprived, limited to four or five hours,
during the eight-hour window where the other
people were actually sleeping, they had to
lie recumbent in bed with no physical movement.
[Rhonda]: Oh, man.
[Matt]: So you controlled, but it was a great
study because you control for physical activity
as well. So we now really understand the link
between a lack of sleep and poor glucose management,
and very very well indeed. And we also know
the stage of sleep that’s important. It’s,
again, deep slow-wave sleep. The study that
I described before, where you’re playing those
annoying tones just below the level of awakening
so I can remove your deep quality of sleep,
you can do that same thing again, and you
essentially produce that same diabetic-like
consequence just by removing or excising deep
[Rhonda]: So you’re doing that, and then on
top of that, some of your work has shown that
you’re now also gonna have a preference to
eat the very food that raises your blood sugar,
[Matt]: Yup, yup.
[Rhonda]: And it’s something I’ve definitely
[Matt]: High glycemic food…
[Rhonda]: Yeah, yeah.
[Matt]: So this is a whole…So it all fits
in, I think, to this energy balance. So, you
know, firstly, what we know is that your regulation
of blood sugar, of blood glucose is profoundly
impaired by a lack of sleep. But what about
your calorie intake? What about your obesogenic
profile, setting aside your glycemic profile?
So what we’ve discovered is that when you’re
not getting enough sleep, two appetite-regulating
hormones go in opposite directions. These
two hormones are called leptin and ghrelin.
And I often, you know, think of them as, you
know, they sound like hobbits to me, at least,
[Rhonda]: That’s hilarious.
[Matt]: Leptin and ghrelin that, you know,
J. R. Tolkien would have written about them.
But they’re not. They’re real hormones. Leptin
is a hormone that signals satiety. What I
mean by that is, when you have high amounts
of leptin, it tells your brain you’re full.
You’re satisfied with your food. You don’t
want to eat anymore. Ghrelin is the other
hunger hormone. When ghrelin is increased,
you feel unsatisfied by your food. You feel
hungry. And you want to eat more.
So after you eat a meal, normally, what happens
is that levels of leptin increase and levels
of ghrelin, the hunger hormone, you can think
about ghrelin as like a grumbling tummy, ghrelin
goes down. And this means that you stop eating,
you don’t want to eat anymore, and you’re
full for a duration of time period. When you
are sleep-deprived, levels of leptin, which
normally signal to your brain you’re full
and you’re satisfied with food, that hormone
is impaired by a lack of sleep. So you lose
the fullness satiety signal in your brain.
If that wasn’t bad enough, the hunger hormone,
ghrelin, actually increases. So it’s a double
whammy effect here. What does that lead to?
It leads to a strong obesogenic profile of
energy consumption, of food consumption. So
what we, and Evan Carter, in fact, has done
all of these pioneering studies, and we’ve
done some of them now, too, typically, you
tend to over eat during main meals. So you
will typically eat somewhere between about
200 to 300 extra calories. If we give you
a meal, and we measure all of the food on
your plate, and we ask how much did you eat
at that one sitting, you tend to overeat by
about 200 or 300 calories per main meal sitting.
[Matt]: Now you could say…Well, actually…
[Rhonda]: That’s after a full night of sleep?
[Matt]: Well, that’s when you’ve been limited
to maybe four hours of sleep for a week. Now
someone, and the reviewers of these papers,
you know, in the early days, said, “Well,
that’s because when you’re awake longer, being
awake is more metabolically demanding, and
you’re probably moving around more.” Not true.
It turns out sleep, firstly, is a remarkably
metabolically active state. You’re very active
metabolically when you’re asleep. In fact,
the difference between being asleep versus
being awake is only about, for a whole night
of sleep, the difference of about 140 calories.
[Matt]: Just like a small cookie, essentially.
So, yes, when you’re awake longer, you do
burn more calories. But the amount of calories
that you increase in your intake far exceeds
anything that you expend by way of being awake
a little bit longer. If that wasn’t bad enough,
though, it’s not the main meal where the trouble
lies with sleep. That’s some of the problem.
The other studies that are even more sort
of devilish are where you give these sleep-deprived
subjects the same meal but then you give them
a snack bar. And it’s an ad-lib food buffet,
essentially. And you can eat as much as you
want. And they allow you to eat in a room
by yourself, so there’s no social pressure,
so you don’t restrict your eating or…it’s
And of course, we measure everything that
you eat. And it’s snacking that is the dead
giveaway here. You end up eating 300 to 400
additional calories by way of snacks. This
is after they’ve eaten a 2,000-calorie meal
in one sitting. They will then go away, and
they will eat an additional 400 calories at
the snack bar.
[Matt]: So, as a consequence, what you’re
doing is you’re increasing the total caloric
intake, which sets you on a path towards being
overweight and obese. Then the discovery came
that it’s not just what you eat…sorry, it’s
not just how much you eat but it’s what you
eat. And what Evan Carter discovered in her
early studies was that if you give one of
these sort of food buffets…and you can eat
anything. It goes from, you know, sugary treats
to salty treats like pretzels or potato chips
to heavy-hitting starchy carbohydrates, breads,
pasta, pizza all the way to salad, what you
find is that you eat more of all of the food
groups. But you eat mostly in terms of an
increase the starchy, heavy-hitting carbohydrates
as well as the sugary foods.
And if we know anything from the recent movement
in food, particularly from people like Gary
Taubes, who has written wonderfully about
this type of stuff, that’s the food that sets
you on a path, again, down an obesogenic and
also a poor glycemic control profile. If you
look at all of the cardio metabolic markers
of health when you’re eating excessive carbohydrate-heavy
rich or and particularly simplified sugars,
they’re bad for your health profile in general.
It’s exactly the foods that you eat when you
[Rhonda]: Is there a feedback loop on those
foods affecting your sleep?
[Matt]: There is.
[Rhonda]: There is.
[Matt]: So I would say this is probably, apart
from maybe the gut microbiome of which there
is now some evidence, and we’re starting to
look at this, too, at the sleep center, it’s
probably the least well understood. But the
bottom line is that if you’re eating a diet
that’s high in carbohydrate, especially high
in processed simple sugars and low in fiber,
you tend to have worse sleep, you take longer
time to fall asleep, the amount of deep sleep
that you get is less, and you have more fragmented
awakenings throughout the night.
Now I don’t think there’s enough evidence
right now for me to hang my hat on any dietary
prescription for sleep. In five years’ time,
I think we’ll have a very different conversation.
I think, you know, absence of evidence is
not evidence of absence. I think we’ll get
there. And I think food is going to be a big
part of that equation. I think we have gut
microbiome, too. I think that’s what we’ve
seen when you, in a few of the studies where
you limit people, you sleep deprive them or
you put them on a jet lag routine, you see
the balance between the sort of bacteriodetes
sort of class of the microbiome versus the
firmicutes. That ratio goes in a balance that
you don’t want.
So, typically, in sort of obese people or
people with diabetes, you get a higher ratio
of the firmicutes relative to the bacteriodetes.
When you modulate sleep and you shortchange
sleep or you put sleep on a jet like profile,
that’s exactly the same gut microbiome sort
of dysregulation profile that you see. So
I think it’s gonna be a great…And I think
[Rhonda]: Because they’re on their own circadian
rhythm as well, the…
[Matt]: They are. Yeah.
[Rhonda]: So is that why? Because their ratio
is changing and then…?
[Matt]: I think it’s possible that it could
be a circadian rhythm. I think it’s also possible
that it could be down to what we described
before, which is the fight-or-flight branch
of the nervous system, that when you are under
slept, it’s, like, revving a car but in neutral.
It’s desperately bad for the engine. That’s
what happens when you’re sleep-deprived. You
get this ramping up of your fight-or-flight
system, and it just stays there with chronic
sleep deprivation for as long as you’re chronically
sleep-deprived. And one out of every two adults
in first world nations is chronically sleep-deprived.
I think it leads to a release of stress chemistry,
particularly cortisol. Cortisol is known in
the gut microbiome to produce this imbalance.
What’s the pathway? I think it’s probably
a bunch of nerve fibers from the brain into
the body called the vagus nerve. And we’re
starting to do a lot of work on this now.
I think the vagus nerve we know has a direct
mainline pathway or provides a mainline pathway
from your brain to your gut. And that’s why
there is a very powerful brain-gut relationship.
I think that’s one of the mechanisms by which
the gut is affected by sleep. Which gets me
excited, because if that’s true, flip the
equation, could the gut microbiome be a path
through which we can reverse-engineer a signal
for better sleep in the brain? I think that
now becomes a parsimonious hypothesis, and
it’s one that I’d like to test as well.
[Rhonda]: There was a really interesting study
looking at the effects of a particular strain.
I think it was the rhamnosus strain that…they
produce GABA in the gut.
[Rhonda]: But GABA doesn’t cross the blood-brain
barrier as far as…I think most evidence
suggests it does not.
[Matt]: No. So the brain has GABA, but it’s
not coming from the…
[Rhonda]: Right, right.
[Rhonda]: But there was some studies that
were speculating that there’s, through the
vagal nerve, because some of the efferent
neurons in the gut were stimulating it, and
they were more, like, relaxed or something.
They were producing GABA in the brain through
the vagal nerve. And GABA affects sleep, right?
[Rhonda]: So it sounds like there could definitely
be some sort of interesting…
[Matt]: You know, that’s one of…I think
the things in anxiety that we see is that
it’s a failure of the brain to dampen down
and basically invoke inhibition throughout
the brain to quiet down these regions.
[Rhonda]: That particular strain of probiotics
was shown to help with anxiety. Small trials,
clinical trials in humans. Of course, the
initial studies were preclinical in animals,
but it has been shown to actually help improve…
[Matt]: It does?
[Rhonda]: Yeah. [inaudible 02:14:04]…
[Matt]: My guess is that, if that’s true,
if you were to look at sleep, I would be surprised
not to see a sleep benefit by way of an anxiety
reduction. I think, I mean, in my lab, that’s
probably one of the most reliable things that
we see when we deprive people of sleep, of
[Rhonda]: That anxiety goes up?
[Matt]: Anxiety goes up. We’ve done the dose
response curve. We’ve looked, you know, hour
by hour by hour. And as soon as you get past
probably about 14 or 15 hours of wakefulness,
anxiety starts to increase. The further you
go into that sleep deprivation period, the
more anxious that you get. It’s a very robust,
[Rhonda]: So what do you think…It sounds
like if you were to give people, like, some
of the main things, you know, the main interventions
they can do to help improve their sleep, a
lot of people are interested in prevention,
a lot of people listening and watching are
really interested, including myself, in prevention.
And it sounds like things that lower anxiety
are obviously important for…
[Rhonda]: Yeah, for improving sleep.
[Matt]: Yeah. I think the prescription I would
have…Well, firstly, I think it’s good to
recognize the sort of those four pillars of
sleep that, you know, depth, duration, continuity,
regularity, but I think there are five actionable
things that people could do tonight to start
getting better sleep. And we’ve mentioned
many of them. The first is darkness. You really
do need some degree of darkness at night to
release that hormone melatonin, which helps
trigger the timing of the healthy onset of
[Rhonda]: And you would say about maybe four
hours before your bedtime, or what would you
[Matt]: Yeah, I would say three to four hours
is the time to start thinking about your light
saturation exposure, certainly in the last
hour before bed. And there are ways that you
can do that you can install software on your
computer if you really have to look at it,
but I would advise against that because I
think computers cause and trigger anxiety.
I think, in fact, if anything, my estimation
right now is that the blue light from those
screens is detrimental. I think the evidence
is favoring a blockade of melatonin and a
reduction in REM sleep.
They did this great study. They took people
with an iPad, one hour of iPad reading versus
one hour of book reading. iPad reading dropped
or blunted melatonin by over 20%. The peak
of melatonin didn’t arrive until three hours
later. This is one hour of iPad reading. They
had less REM sleep, and they…
[Rhonda]: They were reading, like…?
[Matt]: They were reading just a book under
dim light versus reading the same book on
an iPad. When they woke up the next morning,
the people who read the iPad felt more unrefreshed.
They subjectively knew that they had not slept
as well. What was interesting is that when
they stopped the iPad reading, there was a
washout effect that was a blast radius of
reading the iPad. And it’s continued into
subsequent nights even though they’d stopped
reading the iPad.
So I think light is a feature, but I also
think part of the problem with computers and
iPads and iPhones…and I don’t mean to sound
like a prude about this, but they do trigger
anxiety. They are what I would describe as
anxiogenic pieces of hardware. Unless you
are, and if you can do this, please write
to me and tell me how you do this, but most
people get a lot of their anxiety infusion
in part through these devices.
A recent study in teenagers are actually demonstrated
that part of the reason that they were having
sleep disruption by using phones was not necessarily
about the light but was because of FOMO, fear
of missing out, that if you weren’t online,
you would miss out on some key social communication.
And they were suffering because of that, too.
But I digress. My first recommendation is
watch the light saturation, watch the light
pollution. And you can think about it like
that in the evening. So an hour before bed,
certainly, close computers down, try not to
stare at phones, and dim half the lights down
in the house. It really is powerful.
The second is temperature. Try to set your
bedroom temperature to somewhere between about
63 to 66 degrees for most people is optimal,
colder than most people think.
[Rhonda]: Yeah, that is definitely colder
[Matt]: Yeah. And if you get cold feet, it’s
okay to wear socks, but cold it must be. The
third thing is what I would suggest is walk
it out, which is don’t stay in bed awake.
This is one of the mistakes that a lot of
people make. In this era, I think most people,
because of anxiety, were wired and tired.
And one of the problems when you are lying
in bed awake for a period of time, and I think
more than 20 minutes is not great, because
what happens is that your brain very quickly
starts to learn the association between being
in bed is about being awake rather than your
bed being the place of sleep. And so it learns
And so many of the patients that speak to,
they’ll say, “Look, you know, I’m falling
asleep on the couch watching television, and
then I get into bed and I’m wide awake, and
I don’t know why.” And the answer is because
you’ve made the association between your bed
is the place of wakefulness, not your place
of sleep. So the answer is you need to break
that association. Get out of bed, go to another
room, in dim light, just read a book, no computers,
and only return to bed when you feel very
And the first night this may mean that you’re
awake for two or three hours. And it sounds
bad, and it sounds strange coming from someone
like me, but that’s better than staying in
bed awake for those two or three hours. And
over time, gradually, by only returning bed
when you’re sleepy, your brain will relearn
the association that it once had when you’re
a child, and you can relearn it, which is
that your bed is a place of deep sleep and
sound sleep. So that’s the next thing.
I think the final two things we’ve touched
upon a little bit, which is alcohol and caffeine.
Caffeine, many people know, keeps them awake.
It’s a stimulant. It’s what we call psychoactive
stimulant. It’s the only psychoactive stimulant,
by the way, that we readily give to our children
in an unregularized free way, which I think
is a problem. Caffeine has several problems
with it. Firstly, because of it being a stimulant,
it can keep you awake and it makes it harder
for you to fall asleep.
A lot of people, though, will say to me, “Look,
I’m one of those people who can have a cup
of coffee with dinner and I fall asleep and
I stay asleep, and I’m just fine.” Even if
that’s true, we and others have done these
studies, if you give someone a standard dose
of one cup of coffee in the evening, 200 milligrams
of caffeine, the amount of deep sleep that
they have is reduced by 20%. You would normally
have to age an individual by 10 or 15 years
to drop your deep sleep quality by 20%, or
you can do it simply by having a cup of caffeinated
drink or coffee in the evening. So caffeine
is a problem.
The other problem with caffeine is its duration
of action. Caffeine has a half-life of about
six or seven hours. And a half-life simply
means the amount of time it takes for a 50%
of the drug to still be in your system or
50% of it to be cleared. Caffeine has a quarter
life of about 12 hours. In other words, if
you have a cup of coffee at noon, a quarter
of that caffeine is still circulating in your
brain at midnight.
So if you have a cup of coffee at noon, it’s
the equivalent of getting into bed at midnight.
And just before you turn out the lights, you
snick a quarter of a cup of Starbucks, and
you hope for a good night of sleep, you know,
it’s probably not gonna happen.
So the advice would be try to cut caffeine
off around about midday if you can. And even
if you’re someone who, you know, can fall
asleep fine, stay asleep, you should just
know that caffeine can still impact your sleep.
It’s what usually creates the vicious cycle,
by the way, that those people will wake up
the next day, they feel unrefreshed and unrestored
by their sleep, they don’t remember waking
up, so they don’t think it’s the cup of coffee,
but now they start reaching for two cups of
coffee in the morning rather than one. They’ve
got now more caffeine in the system. The next
night, they have even less deep sleep. The
next morning, they wake up even less refreshed.
And now they drink more caffeine. Self-fulfilling
Final one is alcohol. Alcohol is probably
the most misunderstood sleep aid or sleep
drug. It’s what most people usually reach
for when over-the-counter medications have
failed. Alcohol is a sedative. It’s a class
of drugs that we call the sedatives. And sedation
is not sleep. So when you drink alcohol, what
you’re doing is sedating your cortex. You’re
just knocking yourself out. And that’s why
you think that you fall asleep faster.
There are two other problems with alcohol.
Firstly, alcohol will litter your sleep with
many more awakenings throughout the night.
And in fact, some people can even see this
on their sleep trackers even though they’re
not necessarily as accurate as my sleep laboratory
would be. Even with that less accurate measure,
you can still see the impact of alcohol in
sleep. So it litters your sleep with awakenings,
fragmenting your sleep. So this comes back
to the third of the four pillars of good sleep,
which is continuity. It takes away your continuous
asleep, and it makes it fragmented.
The final part of alcohol is that it’s one
of the best chemicals that we know for suppressing
REM sleep alongside marijuana. Now, by the
way, I should say that’s THC-specific. If
you look at CBD, the evidence seems to be
less clear. CBD actually doesn’t seem to be
as detrimental to your sleep as THC. If anything…
[Rhonda]: Yeah, because a lot of people, I
mean, again, this is just anecdotal, talk
about how marijuana improve their sleep.
[Matt]: Yes, yeah. So marijuana…We have
two separate sort of the active from the inactive
component. THC, tetrahydrocannabinol, is the
active component. And CBD is the sort of the
non-hallucinogenic component of that.
[Rhonda]: They’re both in if people are taking
a whole plant here.
[Matt]: Typically, yeah. Yeah, exactly. Now
with strains and drops that you can buy, you
can separate those out, you can play around
with the ratios, and you can get very low
THC to very high CBD. THC does tend to decrease
the time it takes you to fall asleep, but
it tends to make you sometimes wake up more
throughout the night, but it certainly does
seem to block REM sleep.
One of the other problems with THC dosing
of sleep is that you become dependent on that
for sleep. And when you stop using THC for
sleep, you have quite bad rebound insomnia,
which leads to then a perpetuating cycle of
dependency. And I think, you know, I’m never
keen for dependency on anything, be it dependency
on, you know, any, you know, food substance
or anything. I think you should, you know,
be able to generate all of the physiological
pathways and processes that you need, you
know, naturally without dependency.
CBD is interesting, though. There is nowhere
near enough data for someone like me to make
any recommendation. So all I can do is tell
you the very limited data that there is right
now. CBD does not seem to produce the degree
of REM sleep impairment that we’ve seen with
THC. What’s interesting is that low-dose CBD
actually seems to be wake-promoting. So you
actually are sort of forcing yourself more
awake. High-dose CBD seems to help people
sleep more consistently.
I would have thought it would it be the opposite,
that low-dose would have been better than
high-dose, but the opposite seems to be true.
Now we don’t know about the dependency. We
don’t know about the long-term effects. So
I am not going to sit here and give any recommendations
regarding THC use or CBD use. If you’re using
any compound, be it alcohol, be it a sleeping
pill, which are…sleeping pills, by the way,
come with deathly consequences and higher
risk of cancer. Or you’re using…
[Matt]: Yeah, markedly increased risk.
[Rhonda]: Is that because there’s something
directly the sleeping pills are doing, or
because the way it affects your…? What does
it do to your sleep, sleeping pills? A lot
of people take Ambien.
[Matt]: So in the past month, 10 million Americans
have swallowed some kind of sleeping aid,
either a prescription or over-the-counter,
monumental amounts. You know, I often joke
that I think, it’s not really a joking topic,
but, you know, it took George Lucas about
40 years to amass $4 billion in profit from
the Star Wars franchise. It took Ambien less
than 20 months to amass $4 billion in profit.
That tells me everything about the insufficiency
of sleep and the desperate need for sleep
in this modern 21st-century era.
But the problem is Ambien isn’t part of that
same class of drugs that alcohol is. It’s
what we call a sedative hypnotic. It works
on the same receptor, which is the GABA receptor.
Now it tickles the GABA receptor in a different
way to that which alcohol does. But what we
found is that sleeping pills, and I won’t
name any names but including the one that
you described, they are sedating the brain.
You’re not going into naturalistic sleep.
If I look at the electrical signature of your
sleep when you’re on a sleeping pill versus
natural sleep, it’s not the same.
Secondly, what we found is that those sleeping
pills can often come with a grogginess in
the morning and some forgetfulness. Third,
what we’ve found is that in animal models,
and this is work that was done by Markus Frank,
who’s a wonderful friend and colleague, he
was looking at how the brain rewires itself
during sleep. And the brain does, particularly
during the development.
And he has a model in animals, where, if you
sort of put a patch on one of the eyes when
the visual cortex is developing, the visual
cortex shifts over to developing more wiring
to the eye that remains open. And it’s called
the monocular deprivation paradigm of brain
plasticity, and it’s a very well worked out
model of brain plasticity. If you give those
animals some exposure once you patch the eye
to the eye that remains, you drive learning
and plasticity, and then you allow it to sleep,
sleep will strengthen the synaptic connections
that have been made during the day by about
[Matt]: So sleep is almost as powerful as
experience during the day. That’s how sort
of strong and powerful it is. But what he
then did was a study where he dosed those
animals with Ambien. Now it turns out that
those animals slept even longer, if you look
at the data, than the animals who weren’t
dosed on Ambien. And the prediction would
be, surely, they would have as much if not
more of that wonderful brain plasticity. The
opposite was true. Ambien-induced sleep resulted
in a 50% unwiring of the connections that
have been made during day rather than potentiating
That frightens me. Because if you look at
the prescription age of sleeping medications
over the past decade, it’s coming down and
down. I don’t know how long it’s going to
be before prescription medication comes in
to a pediatric realm. And if those data hold
up, it makes me worried. So I’m coming off
topic a little bit, but we know that sleeping
pills are associated with a markedly higher
risk of death, as well as cancer, as well
as your susceptibility to infection, particularly
But returning to alcohol, it is definitely
one of the most, I think, misunderstood drugs.
But what I was saying regarding THC, alcohol,
sleeping pills, if you are using anything
to help you sleep, I think you have to ask
yourself, are you just really treating, you
know, an open wound and not really actually
trying to stitch it up? Because what that
is simply doing is masking a problem that
you’re not dealing with, you know. Why is
And this is not a criticism, I’m sorry if
my voice sounds like that, and it’s not, you
know, I’m so sympathetic to people who have
sleep problems. We see them all of the time
here. But you have to ask why. Why is it that
you’re struggling with sleep? Is it because
you’re of a certain chronotype and you don’t
understand your chronotype and you’re sleeping
at the wrong phase of the 24-hour period and
it’s masquerading as insomnia. Or do you have
too much anxiety in your life and you’re blunting
that anxiety with things like alcohol or THC
or sleeping pills? You know, you’re just kicking
the can down the road. You’re just hitting
the mute button. But the movie is still playing
of damage. Whatever is causing that sleep
disruption, you know, is still there. You’re
just sedating your brain and trying to mask
So I think people who are using those things,
if they’re having sleep difficulties, you
should ask yourself, could I examine my life
and really think more about what it is that
is preventing me from sleeping. What’s good
is that you don’t have to turn to any of those
because there is a non-pharmacological treatment
but is just as effective as sleeping pills
in the short-term. It is completely safe.
And it is more efficacious in the long-term.
And it’s called cognitive behavioral therapy
for insomnia or CBTI.
Based on its efficacious nature, and they’ve
been done lots of randomized clinical control
trials, it is just as powerful as sleeping
pills in the short-term. But when you stop
working with your therapist, and you work
with a therapist for several sessions across
several weeks, when you stop working with
a therapist, you don’t go back to the bad
sleep that you have. You continue on with
your good sleep. Unlike sleeping pills, which
is when you stop them, you have what we call
rebound insomnia, which is that you tend to
go back to the bad sleep that you are having,
if not worse sleep that you are having before
you started taking sleeping pills.
And the danger surround sleeping pills and
their lack of true benefit above and beyond
placebo, which is, if you look at these meta
analyses, in 2015 or 2016, the American College
of Physicians made a landmark recommendation.
They said that sleeping pills must no longer
be the first line recommended treatment for
insomnia. It must be cognitive behavioral
therapy for insomnia because of both the deleterious,
deathly, and carcinogenic consequences of…or
association with…I shouldn’t say cause because
we don’t yet know cause. But certainly they
are associated with higher mortality risk
and higher cancer rates. CBTI, this cognitive
behavioral therapy, must be the first-line
The problem is most doctors don’t know about
it because most doctors get a less than two
hours of sleep education during the medical
curriculum. And this is another crusade that
I am on. I am desperately trying to reach
out to all medical curriculum, not just here
in the United States but in all nations where
you have medical programs. We need a greater
degree of medical education. Our doctors are
not at fault at, you know, not understanding
sleep. They have not been educated about sleep,
So I teach a class here at University of California
Berkeley. It’s usually around 500 or 600 kids.
It’s the Science of Sleep. And they get 25,
hopefully, meaningful lectures about sleep
across the semester. They will have had somewhere
between 10 to 15 times more sleep education
than the standard doctor that you will go
and see. You know, most doctors don’t ask
how much sleep that you’re getting despite
it being this fundamental, you know…Sleep
is a life support system, and it is mother
nature’s best effort yet at immortality. And
it’s a key to disease, and it’s a key to potentially
helping treat or even prevent disease. Most
doctors don’t ask because most doctors aren’t
[Rhonda]: Right. Nutrition is another one
of those areas.
[Matt]: It’s right up that, you know…
[Rhonda]: But thankfully, we have people like
you who are doing phenomenal research and
also communicating that research quite nicely.
I read your book. It was a really great book.
It’s “Why We Sleep.”
[Matt]: Thank you.
[Rhonda]: So if people do want to learn more
about your research, about sleep, about anything
related to sleep or what you do, they can
[Matt]: They can find me, unfortunately, all
over the internet. My handle on social media
is @sleepdiplomat, all one word, so @sleepdiplomat.
You will find me on Twitter. You will find
me sleepdiplomat.com on the web. My center
here at University of California Berkeley
is the Center of Human Sleep Science. If you
just google that, you will find that. And
there’s lots of content. And if you would
like an education on sleep, or you’re just
interested in sleep, or you’re struggling
with sleep, the book which has been out now
and it is out now in paperback, is called
“Why We Sleep.” It is published by Scribner
here in the United States, and it’s published
by Penguin back in the United Kingdom.
[Rhonda]: Well, Matt, thank you so much for
taking some time out of your busy day to speak
with me. I really really enjoyed this competition.
[Matt]: You’re very welcome. Thank you so
much for being such a wonderful sleep ambassador.
You know, I am on a mission. I’m desperately
trying to reach the public because I think
this this message has been the neglected stepsister
in the health conversation of today. But I
can’t do it alone. I need to partner with,
you know, fantastic scientists and public
figures, people like yourself who, you know,
provide this platform and this forum. So I’m
going to anoint you as a sleep ambassador
now as well. So thank you. But just generally,
thank you for this opportunity to reach your
audience and speak about sleep. I really do
appreciate it so much.