Ep150 "Can We Engineer Dreams?" with Adam Haar Horowitz

Can you influence what you dream about tonight? Are you spending years of your life in a world you don’t recall? Can nightmares be manipulated as a therapy? Are dreams sometimes predictive of changes in your health before you become aware of them? Join Eagleman with Adam Haar Horowitz, a neuroscientist and dream engineer who spends his working days trying to help people during their night time.

Episode Transcript

Speaker 1 (00:05):
You're going to go to sleep tonight and you're gonna
have dreams. But what if you could influence what you
dream about? And as far as remembering what you dreamt about,
what if you're spending years of your life in a
world that you never recall. Can nightmares be manipulated as
a therapy? And our dreams sometimes predictive of changes in

(00:27):
your health before you become aware of those changes. Today
we talk with Adam harr Horowitz, a neuroscientist and dream
engineer who is spending his working days trying to help
people during their nighttime. Welcome to Inner Cosmos with me,
David Eagleman. I'm a neuroscientist and an author at Stanford,

(00:49):
and in these episodes we sail deeply into our three
pound universe to understand some of the most surprising aspects
of our lives. So last night I had another dream,

(01:15):
not uncommon for me, that I was about to give
a big talk and none of my slides made any sense.
There was an audience of about four thousand people, they
were already seated, but I couldn't actually remember what I
had agreed to talk about at this conference, and my
computer was acting very strangely, nothing made sense, and I
couldn't move stuff around on the slides the way that

(01:37):
I needed to. So the clock was ticking down to
my entry on stage, and there was a building tightness
in my chest. And then bang, I woke up and
here's my normal life and all my responsibilities. And I
got up and within minutes all of those details dissolved.
The conference, the lecture hall, the computer, that pit in

(01:58):
my stomach, all of it slipped away like I was
trying to hold on to smoke. Now, this kind of
thing where we are fully and emotionally immersed in some
bizarre reality, only to come to realize a moment later
that this was all false. This happens to us every
night of our lives. And that's what we're going to

(02:19):
talk about today. What exactly are dreams about. So let's
start at the beginning. When you close your eyes at night,
the world in front of you disappears. As you fall
into sleep, your incoming senses drift away from you, and
you might think that your brain goes quiet, but that's
not what happens. Instead, your brain becomes a generator and

(02:42):
you find yourself somewhere else. You're running through a building,
or you're talking to someone you know but they have
a different body, or you're in a bad situation that
you need to get out of, or whatever. The key
is that you are inhabiting a reality that feels completely
real until you suddenly discover that it isn't. So what

(03:03):
is going on here? Why does our brain construct dreams?
There are several ideas in the literature, and like many
things in neuroscience, the answer depends on the level of
explanation you are asking about. So some years ago I
proposed a theory about why we dream, called the defensive
activation theory, and the idea, in a nutshell, is that

(03:26):
during sleep, when the planet has rotated into darkness, your
visual cortex needs to stay active, Otherwise, given the brain's plasticity,
your other senses will begin to invade that territory. So
dreaming serves as a kind of internal defense mechanism, keeping

(03:46):
the visual system active in the absence of external input.
And in our published research we find this allows us
to make accurate predictions across different species of animals in
terms of how much of their time they spend in
rapid eye movement sleep. If you're interested in this, please
check out episodes eleven and fifty one. Now, that's a

(04:09):
mechanistic explanation about why the brain shoots activity into the
visual cortex. In other words, it's about why the brain
needs to make sure that the lights stay on at night.
But I've always emphasized that the brain is a storyteller.
So when your mid brain blasts random activity into your

(04:30):
visual cortex, you don't see that as random pixels. Your
brain weaves all of this into a sensory narrative, and
the content of that narrative depends on what is in
your brain, from what happened during the day to what's
been burned deep down into your wiring based on all
your experience in the world. And this is why you

(04:54):
dream about your childhood home, or some conversation you had,
or some fear that you have. And because the connections
are much looser during dreaming, these stories can take off
in bizarre directions. And this is the level that we're
going to talk about today. And this matters because the
content of dreams shapes how we feel the next day.

(05:15):
It reflects what we are processing. Your dream content can
rehearse threats or revisit memories, and sometimes when things go wrong,
it can trap you in loops of trauma, which means
dream content is more than just background noise. We're talking
about experiences that you are having. You've got emotional landscapes

(05:39):
that you're walking through, And given that you spend about
a third of your life in this weird Doppelganger state
of sleep, it's sort of strange that you're having all
this experience but you remember almost none of it. So
what if we could access more of that, and what
if we could at least part shape it. What if

(06:01):
we could engineer our dreams. Now, for essentially all of
human history, dreams have felt like something that happens to us.
They are mysterious and slippery and uncontrollable. But there's a
growing field at the intersection of neuroscience and psychology and
technology that's focused on what is called dream engineering. The

(06:24):
idea is that we might be able to influence what
we dream about and how we experience those dreams and
what we recall. So just imagine being able to nudge
your dreams towards the things that you want to dream about,
like problem solving or emotional healing. Imagine being able to
reduce nightmares by reshaping them. These ideas are being tested

(06:48):
in laboratories and translated into apps, and today we're going
to explore that frontier. My guest is Adam harr Horowitz,
a researcher who has been at the center of this movement.
His work expands from dream inception, in other words, methods
for influencing your dream content, to interacting with your dreams
in real time, to technologies for better recall of your

(07:12):
dream content. Given that so much of your life is
spent in this internal world, what would it mean to
take it seriously? Here's my conversation with Adam horror Horowitz. So, Adam,
tell us about dreams. What is happening when you fall
asleep at night and you have these crazy cinematic experiences.

Speaker 2 (07:35):
I think dreams at night are basically thinking while you're asleep,
And so for folks out there, I think thinking about
your day dreams can be very, very helpful. What's that
thing that happens when you stop paying attention in a lecture?
Why is it you start imagining being on a beach
in Hawaii? Why is it that you leave the place

(07:57):
you are and you go to a whole different world.
What is your brain doing? And so I think day
dreams are the same thing as night dreams. The reason
the night dreams are way, way weirder is because you're
not also processing the world. There isn't all this input
on top of that imagination.

Speaker 1 (08:13):
Totally unanchored from the external.

Speaker 2 (08:14):
Totally unanchored. So a dream is what happens when you're
completely untethered.

Speaker 1 (08:18):
Yeah, And what is happening in the brain when you
make a transition into dreaming.

Speaker 2 (08:24):
Yeah, so there is this wild thing happening when you're
transitioning into dreaming.

Speaker 3 (08:29):
It is a global process.

Speaker 2 (08:32):
So sleep happens, of course in your brain, but your
breath is changing, the way your eyes are moving is changing,
your muscle tone is changing, your whole body is changing.
Sleep is a global, dynamic, complicated process. People sometimes think
about falling asleep as your brain turning off. It is
absolutely not turning off. It is so complex, it is

(08:54):
metabolically super active. Also, there's a lot happening. There's a
lot of calories being consumed, and so when you start
falling asleep and you don't have to process the outside world,
you can start processing internal information instead.

Speaker 1 (09:07):
Yeah, you know, it's like these giant chip making factories,
computer chips, They switch over the whole factory to manufacture
something else. And it's this enormous process where they're switching
everything over. And I think about the transition into dreams
and into sleep more generally as being this giant turnover.

Speaker 3 (09:27):
WHOA.

Speaker 2 (09:27):
Let me give you an example, which is so fun.
When you fall asleep, one of the things your brain
has to do is clear out waste products from during
the day.

Speaker 3 (09:35):
Because all this.

Speaker 2 (09:36):
Thinking, using all this atp, all this identity and triphosphate,
the energy that your cells use, it leaves behind waste
products and you have to clear them out, just like
you have to clear out waste products from your gut.
The distance between your neurons will increase by sixty percent
every night that you fall asleep, so that you can
wash away that metabolic waste. So literally falling asleep is

(10:00):
it's that weird. It's like your brain is expanding to
have these waves of fluid come through and wash you
clean from the thinking of the day.

Speaker 3 (10:08):
It is a crazy, beautiful process.

Speaker 1 (10:11):
Excellent. So let's go into dreaming, in particular, the cinematic
emotional kind of thing. Now, traditionally that's been thought to
happen during rapid eye movement sleep when you can see
when you're watching somebody sleep that their eyes are moving
back and forth beneath their eyelids. But tell us the
modern view about when we're having those cinematic emotional dreams.

Speaker 3 (10:31):
Yeah, so.

Speaker 2 (10:33):
The consensus in the field now is that we're dreaming
all night, We're not just dreaming in REM. And it
is true that if you want those biggest, emotional, most
cinematic dreams, you're most likely to get them in REM,
but you're most likely to get them in REM at
the end of the night. So like the most fun
thing if I tell folks all you want to have
big dreams, snooze that alarm, Why because you're most likely

(10:56):
to have super dense REM right there in the morning.
But also it looks like in those last two hours
of sleep, your non REM dreams look very cinematic, look
very emotional, look very vivid, have first person narratives. They're
hard to distinguish. So I think the big takeaway is
you're dreaming all night, and sleep staging needs to get
a little smarter to tract those dreams. But if you

(11:17):
want to make a good bet and you're watching a
partner a kid sleep and you want to get a
big dream report, and you see those eyes moving, it's
a good time to wake them up.

Speaker 1 (11:25):
Cool, So why are dreams so bizarre?

Speaker 3 (11:29):
Great?

Speaker 2 (11:30):
So I think you can't answer this question without asking
about what dreaming is doing in general. And so I
like a theory called the next up theory, which is
that network exploration to understand possibilities. And this bobstickled and
Tony's odra, I'm too wonderful scientists. And the basic idea
is that, Okay, we go around all day collecting all

(11:54):
this information into short term memory, but somehow that information
has to leave short term memory and get integrated into
the story of our lives. And you can't do that
integration while you're processing the world around you. So you're
too busy taking information in into short term memory, things
that are happening to you right now, prossing the world,
the physics, the possibilities, the dangers. At night, your eyes
are closed at night, you're still you don't need to

(12:16):
be doing any of that work. So what you're doing
at night is taking that information you learn today and
fitting them into the memory networks of your whole story
of your life in particular, so you can predict all
the possibilities of what's going to happen tomorrow and next
week and a month later. So we're having this conversation.
I'll store it in short term memory, but it kind

(12:37):
of reminds me of that time that I got interviewed
to get into high school to have this exam, and
it was kind of intense and emotional. But it's like
a little bit like this in my stomach. It kind
of feels stressful, but it's like not that stressful. It's
like kind of different. I have to understand the similarities
and the difference is, So maybe tonight I have a dream,
and that dream is you and me sitting here, but
you're my high school counselor, and so at night I'm

(12:57):
associating these things largely based on So dreams are largely
emotion driven, and I'm trying to fit these memories together
so I can make sense of the story of my life.
Because our job isn't and you know this as an neuroscientist,
our job isn't to store all the information we've ever
gotten from the world. Our job is to extract patterns.
Extracting patterns relies on making associations those associations are driven

(13:21):
by feelings. This is something that Antonio Dimasio writes about
so beautifully in the feeling of what happens, Dreaming is
the same thing. Dreaming is a feelings driven association to
make sense of your day. And you can do it
because you're not processing all the potholes and hot coffees
and things that are in front of your face while
you're awake.

Speaker 1 (13:39):
Great, and so this is what the field calls consolidation,
where you take all this stuff and you burn it
into the rest of the system. Okay, so the reason
that dreams are so bizarre is because of these weird
associations that are happening while you're stitching this stuff into
what you already know.

Speaker 3 (13:56):
Yeah.

Speaker 2 (13:56):
So the reason I think dreams are so bizarre you
can think about as a kind of network or node association,
a kind of tree of exploration. The first thing you
want to do with this interaction us sitting here together
is think of the most similar memories. Oh, okay, sitting
here is sort of like how I sat in the
chair this morning, and they feel a little bit similar. Okay,

(14:18):
I'm gonna associate those two memories. But the longer you
spend to sleep. The more times you have an altradean cycle,
the more times you wake up and go back to sleep,
the more bizarre those things are going to get. So
that you can understand the most wild and wild and
wild possibilities of what might happen the next day. And
the idea here is kind of simple. It's that you

(14:39):
want to have the widest net of possibilities so that
you're not surprised when you wake up the next day
by what happens. You want to be able to predict
everything that happens tomorrow, and so tonight I rehearse all
the possible scenarios that are similar to this. And the
idea is simply that being surprised costs you a lot

(14:59):
of calories you have to learn. But if you've already
predicted tomorrow how something's going to feel, what it reminds
you of, so that you're not surprised the next time
you sit down in an interview, it basically isn't as
surprising and doesn't cost you as many calories to understand it.
So dreaming is a way to explore short term, long term,
less bizarre and more bizarre associations between a memory and

(15:22):
experience and all the possible experience that it could be like,
so that I can predict them happening.

Speaker 1 (15:28):
What about nightmares? What's the story with nightmares?

Speaker 2 (15:31):
Yeah, So the idea is that dreams, when they're working well,
are a way to integrate today into the story of
my life and make it make sense. And memory consolidation
keeping memories, memory evolution, extracting patterns from those memories, and
just from those memories. Nightmares are when that memory consolidation

(15:51):
evolution breaks. And so the kind of canonical nightmare that
folks in my field worry about are these trauma related nightmares,
and they're often treatment resistant, and so what they look
like is somebody has an experience and that's a traumatic experience.
And these dreams are weird because they're not a metaphor,
they're not a narrative, and they don't change. People have

(16:13):
the same experience at night that they had during the day,
and it's like that memory is not evolving.

Speaker 3 (16:19):
It's like you're going back.

Speaker 2 (16:20):
He was chasing me down the hall with the knife,
just like he did in real life. And every night
I go to sleep and I know that he's going
to chase me down the hall with the knife. So
I actually got into this field because I had a
traumatic experienced little kid, a violent mugging experience, and I
had these traumatic nightmares, and a therapist and my mom

(16:41):
really helped me with it, and helped me by basically
telling me bedtime stories. We can get into how that
works in a second, but I had personal experience these nightmares.

Speaker 3 (16:51):
They happen every night, and so.

Speaker 2 (16:52):
That memory can't evolve, that memory can't lose its emotional power,
that memory can't be integrated into the networks of your
whole life's story. It just returns every night, and it's
like an exposure every night. And those nightmares are really stressful.
You can see that they're stressful all the way from
cortisol levels to them predicting levels of suicidality and depression

(17:14):
and dementia and Alzheimer's. Nightmares are really really, really bad
for you, and so their treatment target. But I think
of nightmares basically as that memory integration system not quite
working and instead getting stuck on the memory because it's
so emotionally charged.

Speaker 1 (17:29):
And let's zoom in on that for a second. About
predicting dementia, tell us this quite surprising result.

Speaker 2 (17:36):
It's really interesting. So This is a research mostly from
a person named Obademi O tai Ku. That takeaway is
that across dementia, across Alzheimer's, across even early death before
seventy five. If you ask people on a set of
questionnaires called a PSQI, which you ask everyone about their

(17:56):
sleep quality, there's this one question, do bad dream make
your sleep worse at night or keep you up at night?

Speaker 3 (18:03):
If they endorse that.

Speaker 2 (18:04):
If they say yes, bad dreams keep me up at night,
people seven years later are more likely to have cognitive decline.
Fifteen years later are more likely to have Parkinsons. They're
more likely to die early. The reason is kind of obvious,
which is that during the day, if you spend a
lot of your time stressed out, we already know it

(18:25):
has all these consequences in terms of inflammation, in terms
of physical health during the day.

Speaker 3 (18:30):
It's not weird.

Speaker 2 (18:30):
We think, oh, my psychological health affects my physical health.
Being stressed out up here changes my whole body. We
know that it's a treatment target. We haven't really figured
out the link between stress at night in dreams and
stress in the body, but it looks like where your
body should be repairing, whether it's your immune system at

(18:51):
night or fighting inflammation at night. Instead you're rehearsing the
worst moments you've ever had and going back to them.
And so there are these consequences in terms of brain health,
these consequences in terms of physical health.

Speaker 1 (19:03):
Is there a way for us to know the arrow
of causality? There?

Speaker 3 (19:06):
Totally?

Speaker 2 (19:07):
So, I mean, the thing you can do is you
can look at confounding factors. So these things that I mentioned, oh,
nightmares and stress dreams predict cognitive decline dementia. Those are
independent of depression, anxiety. Those are independent of sleep quality
and sleep efficiency and sleep duration. You can be sleeping
the right amount of hours, but these nightmares are the

(19:29):
distinguishing factor in terms of whether you are or not
gonna have cognitive decline later on. So one of the
ways you look at is confounding factors. The better way
to determine causality is help with the nightmares, see if
it changes things. So the kind of amazing punchline is
there's a behavioral therapy called imagery rehearsal therapy.

Speaker 3 (19:47):
You take the nightmare.

Speaker 2 (19:49):
Oh, he was chasing me down the hall with the
knife and it was night time and I was running,
and you stop and say, hey, what if it was daytime?

Speaker 1 (19:57):
You ask the patient this during the day.

Speaker 2 (20:00):
Yeah, you're sitting with them and they tell you about
their nightmare that they're having every night. Okay, he's chasing
me down the hall with the knife and it's nighttime
and I'm running. Okay, what if it was daytime? They said, no,
it wasn't daytime. No, no, no, okay, but we're we're going
to try to change it. And you write down, okay,
chasing me, but it's daytime, and then you rehearse it,

(20:20):
and then you repeat it, and then you write it
down again, and you rehearse it, and before bed, they
rehearse it again, and you do it rehearsing.

Speaker 1 (20:25):
They think through that scenario.

Speaker 3 (20:26):
Okay, that's right.

Speaker 2 (20:27):
They really imagine it, really imagine it chasing, but it's daytime.
Do it for three or four nights, and eventually the
nightmare will change.

Speaker 3 (20:36):
They're still getting.

Speaker 2 (20:37):
Chased, but it's the day and they will wake up
and they will say, oh my god, it's not happening
to me. I'm making that dream, I'm making that nightmare.
And so there's this sudden shift in agency what people
call self efficacy or internal locus of control. I'm in control.
I'm not a victim of my nightmare every night, and
anyway it's not happening to me, I'm making it. And

(20:58):
so then off what will happen is people will stop
having the nightmares entirely. But also there's a difference between
having nightmare and being distressed about the nightmare, and saying
I'm in control changes everything about distress. So in terms
of causality, it is very clear that, for instance, helping
people with imagery rehearsal therapy, changing their nightmares, changing the

(21:21):
level of joy and their dreams will predict, for instance,
suicidality a day later, a week later, six months later,
is really important for waking behavior.

Speaker 1 (21:44):
One more question before we move on to the next thing.
I have nightmares occasionally, not very often, but when I do,
they're not predicated on real memories. So how does that
fit with the story that you mentioned about a memory
that's embedded.

Speaker 3 (21:56):
Yeah, totally.

Speaker 2 (21:56):
So I think it's useful to think of these trauma
related repetitive nightmares as pretty distinct from the nightmares that
all of us have sometimes. And then also within that
dream science says a bad dream happens all the time
to everyone. The majority of dreams are negative, actually, like
seventy percent of dreams are negative.

Speaker 3 (22:14):
It seems like we're rehearsing a lot of.

Speaker 2 (22:15):
Threats because we're like, oh, if you're going to rehearse
something that might happen tomorrow, it makes sense to rehearse
the bad thing, because that's the thing that's going to
cost me a lot tomorrow if.

Speaker 3 (22:24):
I don't think about it.

Speaker 2 (22:26):
But these repetitive nightmares seem damaging in a way that
normal nightmares aren't. And normal nightmares, we say a nightmare
is the dream that wakes you up, a bad dream
that you have and you sleep through the night even
if it feels really bad. That's not called a nightmare.
In terms of clinical nightmare disorder, nightmares disturb sleep. Nightmares
disturb your life. They disturb your waking behavior. They make

(22:48):
it so you can't work, they make it so you
can't think, They affect rumination, they return during the day.
A nightmare disorder is distinct from the bad dreams we're
all having naturally.

Speaker 1 (22:58):
So what are the approaches of the therapeutic approaches that
one can take it to nightmares?

Speaker 2 (23:03):
Yeah, I'm just gonna say my favorite recent one so
beyond imagery rehearsal therapy, which is wonderful, and it's a
behavioral therapy that works for many people, but not all.
We need new innovative treatments. This repetitive nightmare complaint is
something you can see in eighty percent of PTSD patients.

Speaker 3 (23:21):
It is huge.

Speaker 2 (23:23):
It is huge if you think only of for instance,
folks were veterans and post combat nightmares, this is millions
of people we're talking about every night nightmares, So big deal.
My favorite recent one is Pillarinsika, Boris Heifitz and Harrison
Chow folks out of Stanford Medical School here actually where
you are, Yeah, who are just great doing this crazy

(23:47):
innovative thing. And what happened was people were coming in
for surgery, a normal surgery that had nothing to do
with their PTSD. They happened to have PTSD, they happened
to have these repetitive nightmares. They go under anesthesia, yeah,
for the surgery propofaal anesthesia in particular, and these people

(24:07):
wake up and they say, oh my god, I just
had the best dream I've had in years.

Speaker 3 (24:14):
It was like my nightmare, but it didn't go bad.

Speaker 2 (24:17):
It was like, this thing that happens to me every night,
except that there wasn't any of the negative emotion. And
they wake up and they say, I haven't had a
joyous dream in years. Thank you for giving that to me.
And then three months later, six months later, two years later,
they no longer endorse PTSD, They no longer have traumatic nightmares.

(24:38):
Just from one session of propofol anesthesia, they are not
having these trauma related nightmares anymore. And so Boris and
the team were like, this is crazy. It keeps happening.
I think that we're going to really investigate this and
we're going to use propofol. And it turns out if
you look at propofol dreams, everyone's having dreams under anesthesia,

(24:59):
and if you look in p particular at the content,
they're about ninety five percent positive and joyous. Doesn't matter
if you come in endorsing PTSD netmores or not. They're
super joyous dreams. And so over somebody who hasn't had
a good time in their head for a long time,
that's a big deal. And so they're exploring this kind
of as a new psychedelic treatment mediated by anesthesia. The

(25:21):
reason this is really cool is because millions of people
are getting in anesthesia today. Yeah, and millions of people
are not getting MDMA treatment today if you think that's
useful for PTSD. But this is something that's already happening,
and all you have to do is change a little
bit of the outset of propofil extended a little bit
and ask them what they're dreaming about. Quietly, don't interfere

(25:42):
with it as they're coming out of surgery, so that
they remember it, because otherwise they'll lose this pleasant experience
if you don't ask them in the right way. Just
like in the morning, if I wake you up with
a loud alarm and I say you got to get
to work, you'll forget your dream. So you have to
help them recall that dream at the end of surgery
if you want it to be beneficial.

Speaker 1 (26:00):
We're going to come back to this issue of dream
recall in a few minutes, and you're also doing some
nightmare work. Yeah.

Speaker 3 (26:05):
I'm really moved by the nightmare problem.

Speaker 2 (26:07):
So we got to do some work with doctor Wesley Jungren,
who's wonderful. He's a marine himself, so he has an
understanding of what it means to be in combat. He
has an understanding from his community of what it means
to take that combat home with you after service and
the ramifications mental health wise, and so Wesley had this
great idea which was basically, Okay, imagery rehearsal therapy beautiful treatment,

(26:31):
but a lot of people don't believe it's gonna work, Like, yeah, right, doc,
you can change my nightmares. Bs, I don't believe you
for a second. So treatment credibility really matters in all
kinds of behavioral therapy. So he said, Okay, okay, I'll
leave you alone at night and you can go to sleep,
but I'm not gonna do anything but just come in
for a nap. And come in for a nap, and

(26:51):
we're going to try to choose something for you to
dream about. Maybe a tree, maybe a fork, maybe Beyonce,
maybe chocolate.

Speaker 3 (26:58):
I don't care.

Speaker 2 (27:00):
I'm lay down on the couch. You use a method
called targeted dreaming gabation, which is something myself and some
friends at MIT came up with. The Basic idea is
you let people slip into the sleep onset period. They're
half asleep, so they're losing a lot of their cognitive control,
but they can still hear. When they're half asleep, you

(27:21):
literally play the words to them from a speaker. Remember
to think of a tree, Remember to think of chocolate.
You then let them slip into the lightest sleep. It's
called covert rem. It is at sleep onset, but it
looks rem like in terms of the dreams. You then
wake them up and say what were you dreaming about?
And they say, oh, in a river, chocolate with umballumpas.

(27:43):
And then you say, okay, you can go back to
sleep now, and you repeat it, repeat it, repeat it,
and you wake them about five times.

Speaker 3 (27:48):
They nap for about an hour.

Speaker 1 (27:50):
How do you wake them gently?

Speaker 2 (27:51):
You wake them gently with really really quiet audio. I've
done vibration, I've done smell, I've done all this stuff.
But simplest is sound, and you just say, hey, tell
me what you are dreaming about.

Speaker 1 (28:00):
Oh, the sound of your voice saying what we're dreaming about?

Speaker 3 (28:03):
That's it. Yeah.

Speaker 2 (28:03):
And so this is via an app that we've made.
This is recorded sounds. So people fall half asleep, remember
to think of a tree. They fall into covert rem
to have a dream. You wake them up what were
you dreaming about? And then you snooze, rinse, and repeat,
do it again, and so then take in one hundred

(28:25):
people about ninety two percent and then will have a
dream of that thing. And this is we've done these
studies in my tea independently replicated a Duke Folks at
Harvard using it Folks at Montre Medical Vancouver General Hospital
is replicated.

Speaker 1 (28:38):
And this is called targeted dream incubation.

Speaker 2 (28:41):
Targeted dream incubation. The targeted Yeah, just because you can
pick a target like tree. So the important thing in
the context of the work that we did with Wesley
Jungreen at the VA is people then wake up from
that nap and they say, oh damn, maybe maybe you
can change these dreams and so treatment credit abilities of interest.

(29:01):
But what we really focused on is these feelings of agency.
And we saw this twenty four percent increase in people's
ratings in terms of feeling of self efficacy or agency,
and then we saw that was related to lower suicidality
the day of and lower suicidality a week later. So
just making someone feel less helpless about their dreams and

(29:21):
nightmares has this real clinical impact. Super cool, and so
all these strategies, imagery, rehearsal, therapy and as she's a dreaming,
targeted dream incubation. We're coming up with a whole menu
for this really important thing.

Speaker 1 (29:34):
Oh incredible, Okay, So let's zoom in on the hypnagogic state,
which is right when you're falling asleep. So some people
like Thomas Edison and Solvador Dali would actually use this
take advantage of this. So tell us how they did.

Speaker 2 (29:51):
That, totally, So pick your Edison's or your Dolli's. But
also you can read Proost, and you can read Nabokov,
and you can read all these poets and and scientists
playing in this little area of sleep on set the Beatles,
Einstein and whatever. So it's the moment when your brain
is turning thoughts into images. It's the moment where your

(30:12):
brain is turning the outside world into an internal world.
You're awake, you're asleep, but it's not a switch. Sleep
scientists think there are nine stages of the descent into sleep.
They're called hory stages, and they're quick.

Speaker 1 (30:25):
How do you spell it?

Speaker 3 (30:26):
H ori I.

Speaker 2 (30:26):
It's named after Goud, named to dow Horri. There's a
bunch of amazing Japanese dream researchers.

Speaker 3 (30:31):
Let's all about it.

Speaker 2 (30:32):
Man In those nine stages of sleep. You have these
kind of different stereotype dreams, you have these very specific
brain wave features EEG features, And Hypnagotya is the state
of transition. The kind of cool thing about it is
you're neither here nor there completely, So for people listening,

(30:53):
this is the state when you're falling asleep and you're like, yeah,
I've got to go to the grossery store and I've
gotta picked my kid up from school, and I've got
to be in a bouncy castle with my grandma. There.

Speaker 3 (31:08):
Wait no, wait, wait, wait, it's where.

Speaker 2 (31:09):
You're It's where your thought takes a left turn all
of a sudden, and it's quick, and it's very very
very strange, And if you want to play with it,
you can extend hypnagadja and it will only get weirder
the longer you stay in it.

Speaker 3 (31:22):
It is a crazy state of mind.

Speaker 1 (31:24):
Hmmm, how do you extend it?

Speaker 2 (31:25):
So I extend hypnagadia with serial awakenings that sleep on
set so very very simply, you can, for instance, without
any sort of brain imaging, if you just snooze ten times,
especially over a nap, because naps are going to be
very very REM like during the day.

Speaker 1 (31:41):
As in hit your snooze button on your larm.

Speaker 2 (31:43):
You just keep hitting that snooze Those dreams will get
weirder and weirder, but you will also get more control
over those dreams. So we haven't gotten to talk about
lucid dreaming yet. But lucid dreaming is this idea that
you could in a dream, know that you're dreaming and
then choose what to dream about. And so this is
something that's been documented for hundreds of years but actually

(32:03):
was researched again at Stanford, most seriously by a guy
named Stephen Leberge about fifty years ago. And it's the
idea that in this covert REM early sleep period or
in REM, you realize you're dreaming, you take control of
the dream, You make a choice. But even crazier, if
you are awake inside a dream, you're totally paralyzed. Right

(32:25):
of course, you can't move your body, or else you
would wake up and start running around the bed and
fighting ninjas. But your eyes are not paralyzed in REM
sleep they can move around. So in that lucid dream
the experiments work like this. Somebody realizes a dream and
you've agreed before Hey, when you realize it's a dream,
move your eyes left and right in the dream. Look

(32:45):
at the draft, Look at the elephant. Look at the draft,
look at the elephant. Your real eyes of the person
lying there will actually move left, right, left, right four times.
So you can signal from inside the dream, I know
that I dream.

Speaker 1 (32:58):
Say four times. You mean that was the pre agreed
upon the number to do. How crazy is that? So
they are completely in a different world. They are asleep,
we're on different planes of consciousness. But they're looking around
in their dream one two, three, four. Their actual eyes
are moving one to three four. We register that on
what's called their eog Looking at their eye movements, it
looks like that's right totally.

Speaker 2 (33:20):
And then you can go a little crazier. Okay, they
know they're in a dream. I'm going to play a
math problem on the loud speaker. What's three minus one?
In their dream? They'll be walking down the street and
then they'll look at a house and it'll be house
number three minus one. And let's say, oh, it's the
math problem I knew I would get. And then they'll

(33:42):
move their eyes one two and they'll answer correctly. This
is a literal this is this is this is a
paper kind of Karen Conkling. A bunch of folks led
this as a few years ago, replicated across four different labs.
They are asleep. They are in a different plane of consciousness.
You are playing math problems over the loud speaker. They're
answering correctly with their eyes while they remain in rem
and this is like, it's cool science, but it's also

(34:04):
the first time humans have ever communicated across these planes
of consciousness. So as someone who's really interested in consciousness
and phenomenology, like for you this much, it's pretty weird.
It's like we're we're talking across worlds. So I really
love it.

Speaker 3 (34:18):
Via the eyes, I love that.

Speaker 1 (34:20):
Okay, so that's loose dreaming but zooming back out now
to this hypnogogic state right when you're falling asleep. So
you're about to tell us what Edison Dalli and all
the others actually do to take advantage of this. It's true.

Speaker 2 (34:33):
I got a little excited about dream research. I took
a left turn, gotten to lose babe. So yeah, they
were just something really simple, which they called the steel
ball technique or sleeping with a key. But both involved
sitting up right like you are letting yourself take anap
during the day. Especially good is what's called the post
prandial period, which is a fancy way of saying after
you eat lunch, because you really get sleepy and you

(34:56):
fall asleep sitting up, but you're holding something heavy when
you start falling asleep. In hypnagodia, all these things are
happening in your brain, but also you're losing muscle tone,
so your grip will loosen, you'll drop that ball, it'll
hit the ground and make a loud sound. It'll catch hypnagodia.
Edison called it his genius gap because hypnagodia is super associative,

(35:19):
super exploratory, super creative, super flexible cognition, and so Edison
would pick a problem he was working on, think about
it as he was falling asleep, and catch his hypnogogic
thoughts around that topic to find his most creative thinking.
And totally cool that Edison did it.

Speaker 3 (35:35):
Interesting.

Speaker 2 (35:35):
It's been documented, but we just got real scientific evidence
for it, led by Celia Lecrow a wonderful paper called
sleep onset is a creative sweet Spot, And what she
found is that have people try to solve math problems,
specific kind of math problem that needs a Eureka moment,
like solving puzzles. Have them do this exact technique and

(36:00):
they enter into end one and they wake up, Or
have them sleep a little deeper and let them go
into end two a different stage of sleep. The people
who just dipped into end one have a three x
greater likelihood of solving that creative problem than the people
who dipped into end two and went deeper in. There
is something sweet happening in ND one. There is something

(36:23):
strange and specific and flexible and creative, and I know
you've written about creativity. There is something associative happening in
that state of mind, but you got to catch it
if you want it. So that's what the steel ball
was for, and that's actually what our work at MIT
around Dormea, which was basically a wearable device which would
track the same thing you losing muscle tone, and then
would talk to you. It's really similar to the steel

(36:44):
ball technique, except instead of waking up entirely, it would
help incubate that dream of your problem you wanted to
work on right at that moment. It would wake you
up automatically and record a verbal report, and then it
would say, great, you can snooze again because the ball.
It's kind of hard to snooze again with a steel ball.

Speaker 1 (37:12):
Okay, great, So now let's talk about dream recall. So
what happens with almost all of us is wake up
in the morning, we think, oh my gosh, I had
this dream, but this happened, that happened, and then it's
like gossamer and within fifteen minutes we can't remember it
at all anymore. Yeah, So, first of all, why do
we not recall our dreams? And then we'll get into

(37:33):
how we might totally. So there's a few things that
we do every morning.

Speaker 2 (37:37):
That make it pretty likely we're not going to recall
our dreams, which is, we wake up to a really
loud alarm, which doesn't let us leave sleep gradually, it
makes us leave sleep suddenly we start thinking immediately about
what's on the phone, that email the New York Times,
Oh my god, what is he doing?

Speaker 3 (37:53):
Now?

Speaker 2 (37:54):
All of that is going to invade our thoughts and
overwrite these really gentle bits of dream recall. But there's
another reason we don't recall our dreams, which is we
don't care about our dreams.

Speaker 1 (38:05):
So in this evolutionarily yeah, yeah.

Speaker 2 (38:08):
So in the same way, if I decide, you know,
birding is really important because knowing what birds are around
my city will help me understand climate change, I will
walk out of this recording studio today and suddenly I'll
see birds all around me, and I'll remember those birds.

(38:29):
I decided they were important, so I noticed them, so
I remember them. You have to make things important if
you want to encode them, consolidate them, and keep them.
We don't care about dreams culturally, but I can tell
you if somebody's in a dream study and we say, hey,
we're investigating this thing, it turns out dreams.

Speaker 3 (38:46):
It's true.

Speaker 2 (38:47):
It turns out dreams and specific kind of dreams are
pro dromal. So before a syndrome, they can predict, for instance,
a manic episode or a depressive episode. So we really
want to keep track of your dreams Suddenly, dream recall boom,
it spikes because he made it important. And so if
tonight you go to sleep and you say, the first

(39:08):
thing I'm going to do in the morning is I'm
going to record my dream, and then I'm going to
share my dream with my wife and she expects me
at the breakfast table to have this conversation with her.
Because Mark Blackgrove's research shows that sharing dreams creates more
marital intimacy than sharing waking thoughts.

Speaker 3 (39:23):
That's true. Because of that, I'm going to keep it.

Speaker 2 (39:26):
Or because there's research which shows how useful sharing your
dreams with the therapist is in terms of introspection. Pick
your reason your dream recall will spike. So these two things.
How we wake up in the morning too stressful, too invasive,
you move too much? How we go to bed at night.
I don't care about dreams. Why would I remember them?
You can set salience, you can set meaningfulness of dreams

(39:48):
before sleep, and you will completely change your dream recall.

Speaker 1 (39:52):
Mm let me just return to this question about why
under normal circumstances we don't remember our dreams. Because one idea,
of of course, is that it's a bunch. You know,
you're sticking your head in the night blender and you've
got all these activities happening, But you don't want that
to actually interfere with your memories of how the world works.
You don't want that to actually build your internal model

(40:12):
of the world. How do you think about why generally
we don't remember unless we put a lot of effort
into it.

Speaker 3 (40:17):
I think it's really hard.

Speaker 2 (40:19):
And I know you like to think of things both culturally,
socially and evolutionarily. I think it's very, very hard to
tease these two things apart. The reason being, you know,
I've been super, super lucky to spend some time in
indigenous context, for instance, where dreams are deeply important to people.
Dream recall there, everyone's remembering dreams every day. I think

(40:40):
there's a cultural question as much as there's an evolutionary question.
And I think the big reason we don't remember dreams
isn't because of oh, necessarily there's some way that our
brain works. I think it's because we don't care. I'll
give you another example. Canonically, there's no smell in dream reports.
I expect this is the case with you. Oh, I've
really primarily visual dreams. I have a lot of motor

(41:03):
I run around, things happen, people attack me.

Speaker 3 (41:06):
But I'm not smelling stuff in my dreams. That's not
why I wake up reporting.

Speaker 2 (41:10):
But again, if you go to a space where smell
is incredibly important, specifically a tribe called the Aungies and
the Andaman islands where smell is a way of interacting
with your world and spirits and ancestors. The ajrees majority
of them their dreams are smell. I think it's a
dream is related to the thing you think is important

(41:30):
about your world. And we live in a very visual world.
Because of that, we have very visual dreams. Because of that,
we have visual activity in dreams. Because of that, we
can even read dreams out in the visual cortex, and
we can predict what someone is dreaming about really well.
But I think it's a cultural question there is You're
totally right. Lots of people in the memory field think, oh,

(41:52):
dreams are so bizarre and weird. They do their work
and we don't need to remember them for them to
do their associative work. But I don't think that totally
explains it, because there are these other groups of people
who are recalling their dreams every day, and they don't
have different brains than us, they have a different culture.

Speaker 1 (42:08):
Hmm. I take that point about dream recall. I'm slightly
more skeptical about the smell part only because when we
look at what are called PGO waves, which are from
ponting to geniculate to occipital cortex that's why it has
this heavy name. But the point is it's the set
of waves driven by a very evolutionarily ancient part in
your midbrain that blasts activity into the visual cortex, and

(42:33):
only the primary visual cortex. So that's why dreams are visual.
Here's the thing they It splashes out from the visual cortex,
and maybe if you're a culture that emphasizes the smell
a lot, then certain things that are associated, so it
follows some pathway and triggers a smell.

Speaker 3 (42:51):
One hundred percent.

Speaker 2 (42:51):
You're totally right, and I'm stoked that you know about
PGO waves, so on the on the PGO wave thing, totally.
But the thing you need to answer that question is
a brain imaging study of the ages and the Niman
islands to see actually what their sleeping brain looks like
and if it's the same or if it's different.

Speaker 3 (43:06):
So totally, maybe all cool.

Speaker 2 (43:08):
We have PGO waves and we have visual cortex, then
we have ol factory bulba activation, and we can see
the kind of generation of their dreams. Totally, maybe they
sleep really differently. I don't know, it could be pretty interesting.
I see you don't buy it.

Speaker 1 (43:20):
I don't buy it only because we radiated out of
Africa and moved to different places like yesterday. But the
human brain hasn't changed in that time. So and in fact,
these pgo waves are conserved all through the animal kingdom.
Everybody's dreaming this way totally. But I do think I

(43:42):
do take your point that if smell is really important
to your culture, then you'll have more paved pathways between
visual cortex and olfactory and so maybe that'll get triggered
more and you.

Speaker 3 (43:53):
Can do fun things with that.

Speaker 2 (43:55):
You can say, oh, you know, I really want to
make new musical album, and I know that the Beatles
had their best songs come in dreams, and I want
to use my dreams creatively, and so I'm gonna fall
asleep listening to music from this album that I'm starting
to create, so that it's more likely that I have

(44:16):
a dream which is sound based, so that it's more
likely that I have a creative breakthrough in my dreams.
So you can do some kind of picking around this.
You can pick, for instance, oh, I want smell as
a signal that influences my dreams, So I'll give you
a really crazy study.

Speaker 1 (44:33):
Guy.

Speaker 2 (44:33):
Well, I'll give you two smell studies. I'll see if
you like them. They're very very different. One of them.
You can pick smells. I smell them. I smell them,
I spray them. I associated with lavender or something, lavender rose.
I associate them with a place, and then I spray
that smell on my pillow. I'm much more likely to
dream of that place. You want a dream of Paris,

(44:53):
spend the next week. I'm gonna watch a little Paris video.
Wallacecent is in the room. I'm gonna spray that same
scent my pillow. I feel very confident you will have
a dream in Paris.

Speaker 1 (45:02):
I thought these studies were done a long time ago,
so long ago, and okay, but I maybe I misremembered.
I thought they hadn't yielded positive results.

Speaker 2 (45:10):
Yeah, so the study that I'm referencing is two hundred
years old. It's a guy named Saint Denise who would
go with the piece and he spray sense, and then
he would spray the one his pillow. He would go
to the place, and then people have had some trouble.
How does it work, this association? But we're getting better,
and we're getting better, and we're getting better at it.
And I'll give you a really, really specific example of
this protocol, which is basically, I'm going to associate a signal,

(45:33):
it doesn't really matter if it's smell or sound with something.
Then I'm going to reintroduce that thing in sleep. Then
I'm gonna hope that somebody dreams of that thing, and
then i'm gonna see how it changes their next day.
So there's a paper from this year again from Karen Conkly,
whose work I love. You have people look at a

(45:53):
set of puzzles while they're awake. You then, while they're
looking at this puzzle versus that puzzle, play sound A
versus sound B. You then wait until they're in remsleep.
You then play sound A versus sound B. What you
can see is that playing sound A, about seventy five
percent of people will have a dream of the puzzle

(46:13):
related to that sound. Not only will they have a
dream related to it, but they'll then do better at
solving that puzzle than folks who didn't have that reactivation
related to it. So we're getting really good at not
only making a claim, oh we can associate a sound
with something in waking life and then make you dream
about that thing, but making you dream about that thing

(46:35):
will then make you better at tasks related to it.
We did the same thing with creativity. Make someone dream
about X, test them on X. Someone else dreams about
why test them on X. The first person will do
about forty three percent better on a set of creativity tasks.
And it is hard to do causal work and consciousness.
You have done this for a lot longer than me.

(46:56):
But we're getting close. Dan Dennett wouldn't be satisfied, but
we're getting close.

Speaker 1 (47:00):
Yeah. Why wouldn't he be satisfied?

Speaker 2 (47:02):
Because I think that some people think you can't solve
the epiphenomenon problem. You can never make a claim about consciousness.
Even if you say the experience seems to change waking behavior,
somebody would say it's not the experience, it's the neural
mechanisms that create the experience. I think there's basically no difference,
And like, that's an argument for philosophers. What we're supposed

(47:24):
to do is make useful stuff great. Okay, So that
was back to dream incubation. But the other end of
it that I wanted drill in with you is about
dream recall. Okay, so you've developed an app called dust
tell Us about that totally can I say one thing
before and thank you for asking about it.

Speaker 3 (47:44):
So on dream recall.

Speaker 2 (47:47):
The best sci fi scenario is, actually, we don't have
to wake people up and ask them about their dreams
because self report is unreliable.

Speaker 3 (47:55):
The sci fi scenario is what.

Speaker 2 (47:57):
If we could just look at their brain activity and
tell what they were dreaming about. Great in the same
way that sci fi predicts science in so many fields. Horikawa,
wonderful researcher, has this paper and encoding visual imagery during sleep,
and he shows that what you can do is you
can show people bunch of pictures, flash them really really
really quickly while they're awake, and look at their visual cortex.

(48:18):
This is an fMRI paper. It's a functional magnetic resonance
imaging paper, so a kind of brain imaging where you
look at brain blood flow and oxygenation, impress of the brain.
But the very simple idea is flash a bunch of
pictures of people while they're awake, let them go to sleep,
look at their brain activity in the visual cortex, and
look at similarity.

Speaker 3 (48:36):
Ah, when they looked at a.

Speaker 2 (48:37):
Giraffe while they were awake, this part lit up looks
like that part is lighting up when they're asleep, and
you can guess with really solid accuracy what they are
dreaming about.

Speaker 1 (48:48):
And it's it's because you're looking across the whole visual
cortex and you're seeing patterns, and so right, giraffe is
represented by a particular pattern across all these vok which
are three dimensional pixels in the brain. You know, some
are very active summer lots in between and anyway, you

(49:08):
can look at this and that's how you figure out
these things, right.

Speaker 2 (49:11):
Yeah, and you can now do this. There's a paper
from this year actually from Monica Schenhauer, which is doing
this just with EEG. So epher Mariah. Is this big
magnetic million dollar machine, scary, intense expensive EEG. Pop it
on your head with a bunch of electrodes electron cephalogram.
This is the classic brain waves. It looks like a

(49:32):
heart rate based like Doom, Doom Doom. You can put
people to bed, have them listen to one of four
bedtime stories. You can then look at their brain activity
while they're listening. When they go to sleep, look at
their brain activity and you can guess which story they
listen to. So it's getting reactivated just on EG and
the stronger the similarity to the story, the more likely

(49:55):
that they are dreaming of that thing with condition blind
raiders saying, Oh, this person, just looking at their dream,
can you guess what story they listen to? Just looking
at their brain activity, can you guess which bedtime story
they listen to?

Speaker 3 (50:08):
And so we're building up.

Speaker 2 (50:09):
To this stage where we can think, huh, not only
can we control what someone dreams about with bedtime stories,
we can then see them dreaming about it. With brain imaging.
We can then tie that reinstantiation of the experience to waking, performance, encoding, consolidation, creativity.
We're building the IO, the input output system for dreams,

(50:32):
and it's totally, it's totally. It's crazy. It's like a
crazy moment in the field.

Speaker 1 (50:36):
Yeah. Okay, so now let's return to dream recall. Okay,
so you wanted to help people get better at recalling
their dreams because otherwise, unless you live on some island
somewhere where that's important, it's very easy to not recall them.
So what did you do? What is the dust app? Yeah?

Speaker 3 (50:53):
Thanks for asking.

Speaker 2 (50:54):
Yeah, so the dust app says at night, going to
have classes on dreams and tell you why they're important.
In particular that then turn into practices so you can
try incubating a particular dream. So we have classes for
flying dreams. You want to have a flying dream, give

(51:14):
me fifteen minutes before your bedtime. For the rest of
the week will give you a flying dream. We have
classes on image rehearsal therapy. If you're struggling with nightmares.
We have insomnia classes, lucid dreaming classes, yoga nidra classes.
But the whole point of the classes is basically to
make you think that your dreams are important, so overnight
you're more likely to encode them.

Speaker 3 (51:32):
But then in the.

Speaker 2 (51:33):
Morning use our alarm instead of your native Apple iOS
alarm Doom. It'll wake up really gently and it'll say,
tell me what you're dreaming about. You don't have to move,
you speak out loud, you stay still. It'll record your
dream report, store it in a journal, and then ask, Hey,

(51:55):
do you want a snooze or do you want to
wake up? And you can snooze. After that snooze, you
can choose a dream incubation. Hey, remember to dream of flying,
so that when you're half awake, half conscious, not quite there,
it'll guide your dreams towards a particular theme. So it
takes both the targeted dream incubation protocol we talked about,
takes the targeted memory reactivation protocol we talked about, and

(52:16):
it squeezes it into an app so you can use
it at home, use it for your nightmares, use it
for your creativity. We're seeing big jumps in dream recall,
which is really really fun because, like you said, if
you don't remember these things, you can't work with them.

Speaker 1 (52:29):
And when you record your dream content, what happens with
it then.

Speaker 2 (52:34):
Totally so it lives in a dream journal. And then
you have choices. You can analyze your dream. You can
look at what gestalt Freudian or Buddhist approach to it,
or you can look at what a more contemporary cognitive
neuroscience approach to understanding that dream with and that's AI.
That's done by AI. You can also track over time, Hey,

(52:57):
who am I dreaming of?

Speaker 3 (52:58):
Most? What places in my dreaming of?

Speaker 1 (52:59):
Most?

Speaker 2 (53:00):
What stress dreams am I having? You can also look
at your dream readiness score, so in the same way
you can be like, man, you didn't get much sleep
last night, You're probably gonna be really sleepy tonight. You
should be really careful about your bedtime. You can also
look at, Oh, you had pretty disrupted rem in the
night before. We think you're gonna have a super vivid
dream tonight. I think that you should make sure not

(53:21):
to drink a glass of wine and stay away from
the blue light so you can have a big, beautiful
experience tonight. So you can look at your dreams over time,
you can look at your dream readiness, and the dreams
over time matters because across all these different symptoms, pick
your motor symptoms in Parkinson's, or your bipolar disorder, or
your depressive episodes. A sudden change in dreams is a

(53:43):
clinical signal that's super meaningful. And so if you see
this sudden change in dreams, or you see this sudden change,
for instance, in acting out your dreams, it's really something
you should know. But you're not really gonna know that
if you haven't been tracking your dreams over time and
suddenly they're all really violent for ends, or suddenly they're
all dreams of death, for instance, and so useful for

(54:04):
a psychiatrist, for instance, if they're working with someone who
might be struggling with suicidality. Oh, there's this sudden change
dreams related to death. We're going to be really careful
in the coming months, etc.

Speaker 1 (54:15):
Etc.

Speaker 2 (54:15):
So it's a way to understand your dreams over time,
to understand yourself.

Speaker 1 (54:19):
Might it have something to do with biology. I don't
know the answer to this, but if you get a
viral infection, might your dreams change? Yeah?

Speaker 3 (54:27):
It's awesome question.

Speaker 2 (54:28):
So there's so much different work on this, but maybe
the closest work to that kind of question is the
work around lupus. So before you have a flare, before
you have your first flare, often people will have this
sudden change in dreams the kind of pain they're going
to experience a few months later. They'll have it in
their dreams before they have it in waking life. These

(54:51):
are called pro drums. This is not that surprising. It
sounds a little weird, but basically the idea is that
you're having changes already in your body, but they're very
subtle and they're overshadowed basically by all the things you're
processing in wake. But at night you're much more in
touch with your body because you're not overshadowed by input,
and so that little little bit of pain or little

(55:14):
little bit of change is going to manifest as a dream.
So if you're having a lot of dreams of your
teeth falling out, which is a classic, it predicts you
having some dental problems six months later. This is not
that surprising. If you want to have some really messed
up dream tonight, just wait until eleven PM and eat
really spicy food. Your stomach is going to mess your
dreams up. Like we live in a brain that is

(55:35):
connected to a body. It's just like those two things
aren't disconnected. That's sort of surprising to people because dreams
are like ethereal and untouchable.

Speaker 3 (55:42):
They're not.

Speaker 2 (55:42):
They're coming from your stomach. So in the prodrome space,
that means my girlfriend has lupus. So I've thought a
lot about this. That means that, huh, you have this
sudden change, your doctor should know about it. I have
some folks I'm related to a struggle with Parkinson's. That
means that if you have a sudden change in your dreams,
they're suddenly more violent. You're in your fifties, all of

(56:03):
a sudden, you're having these dreams where you're getting attacked
and you're kicking and you're fighting. It's likely that those
dreams So those dreams predict onset of rem behavior disorder,
which predicts on set of Parkinson's. If you know that
Parkinson's is coming, this is decade, this is fifteen years out.
Take that to your collinsion. That is so valuable. So yes,

(56:24):
you asked me, oh, maybe if you have a viral
infection or a fever, could those little changes in your
body manifest as dream content? Definitely, all the way from
tooth changes to a dopamine disorder.

Speaker 1 (56:35):
Amazing. So let me return to two things. One is
the curiosity question. I often hear people say to me, oh,
I have the exact same dream every night, and I've wondered,
how would you actually know if that's true? Because often
when I have a dream some bizarre thing, I think, oh,
I've been here before, but it's presumably not true that
I've had that same dream. So if you're looking at
a dream journal, you could actually answer the question what

(56:57):
do people find?

Speaker 2 (56:58):
I love this question because dreams mess with your feeling
of familiarity. Beyond creating some strange visual world. I mean,
as brains eientists, we have to think of how weird
it is that my mother can be a chair in
a dream. And I'm like, yeah, she's a chair. She's
always been a chair. Something about familiarity is all messed
up in a dream, And so the idea that you

(57:18):
could step into a dream and be like, oh, this
is familiar, I've been here before.

Speaker 3 (57:22):
Well, I don't.

Speaker 2 (57:23):
Really believe you totally like you said. So if you
dream journal over the course of months, you might find,
for instance, you're having a dream of your boss yelling
at you a lot. That's probably something that you want
to work through during the day. You're really returning to
this place a lot. Let me give you a more
concrete example. One of the dust users is in her

(57:44):
seventies and she's bita tester and she's using the app
and she has not been recalling her dreams before she
starts using the app. She starts remembering her dreams, and
she sends this report into the team and she says,
you know, I've been waking up in the night with
my heart pounding for so long, and I don't know
why I'm remembering my dreams suddenly, and I'm having this

(58:07):
dream almost every night. It's about my parents, and my
parents are yelling at my brother and I'm a kid,
and it's sixty years ago, and they're being cruel to him. Basically,
she goes back to a difficult moment in her childhood.
Every night she wakes up from it heart pounding, but
she didn't remember the dreams until now. Suddenly she recalls
the dreams and she writes in in her report, this

(58:28):
is some anger that I think is actually with me
during the day that I was pushing down that I'm
going to take to my therapist and I'm going to
work on it and I'm going to process. But I didn't
remember the nightmare before, and so I didn't know that's
why I was waking up at night. And so dreams,
you know, whether you're a psychoanalyst or your Salvador dull

(58:48):
Lee or your prooster, you're at the temple of a
sclape beyond an ancient Greece. People are saying the same
kind of thing, which is dreams are a place where
more subtle signals can surface than our introspection during the day.
We are more in touch with ourselves and more in
touch with the world, and we're more aware of quiet
and small signals, and so we can really see what

(59:10):
it is we're processing, really see what it is that's
important to us, and so tracking of dreams over time,
whether it's a nightmare like this user, whether it's a
stress dream of your boss, or whether it's you know, man,
I'm I'm having a lot of dream of Instagram ads.
I should probably stop scrolling a bed and falling asleep.
So this is crazy. Fifty This is a survey study

(59:32):
across one thousand people. Fifty four percent of young Americans
eighteen to thirty five regularly dream of advertisements. It's pretty
crazy and dark, but the reason is they're falling asleep
scrolling on their phones and they're getting hit with these things.
And the CEO of Netflix says, my main competitor is sleep.
They are trying to push further and further into the

(59:53):
sleep onset period. So they're basically doing targeted dream incubation,
but with the most sticky, million dollar or content you
can think of. So people are dreaming of ads. If
you dream journal and you're having like a bunch of
Haley Baber dreams, you should probably be like, I probably
shouldn't do this at night. I probably should be doing
more of my own work than letting them get me.

Speaker 1 (01:00:14):
Okay, So I'm closing this part tell us why dream
recall matters totally.

Speaker 2 (01:00:19):
It's sort of like asking why thinking matters. Thinking is
where I figure out who I am. Thinking is where
I figure out what's important to me. Thinking is how
I make my choices about what to do tomorrow. Dreaming
is your thinking at night. It is your most private
and personal space. It is a world made entirely by you.
So it's an important place to introspect. It's an important

(01:00:40):
place to reflect and find yourself. But just the simpler
reason is that you're dreaming in all stages of sleep.
You're sleeping for twenty six years of your life. If
you decide a dreams don't matter, I'm just gonna let
them go, you're giving up years of lived experience where
you could be flying, you could be sitting with a
law loved one, you could be sitting with your grandma

(01:01:03):
across the table. You could be returning to childhood memories,
you could be choosing joy, and instead you're like, oh,
I'm just gonna lose that huge part of myself. And
so one of our users wrote this really beautiful thing.
We said, I feel like I've been sleep walking through
sleep for my whole life until now. Suddenly getting dream
recall means you find a new part of your mind.

Speaker 1 (01:01:24):
Is the unpredictability of dreams something that's useful. If we
were to engineer our dreams one hundred percent, would we
be losing something totally?

Speaker 2 (01:01:32):
I think it's an awesome question. It's an ethically important question.
I think there's no chance of engineering your dreams one
hundred percent. In the same way that, like, you know,
what is education if not trying to engineer people's thoughts
one hundred percent? It is really really hard to put
a thought.

Speaker 3 (01:01:47):
In there the way you want.

Speaker 2 (01:01:48):
You know, you're planting a seed, but in the soil
of their mind. I mean, you're you're raising two kids.
It's like you're planting a seed, but that kid's a garden.
They're gonna grow into something else.

Speaker 1 (01:01:58):
Man.

Speaker 2 (01:01:59):
And so I think dreams are dreams are like that.
I feel totally comfortable.

Speaker 3 (01:02:01):
You come to lab. I'd make you dream of a tree.

Speaker 2 (01:02:05):
But is it the tree from your childhood or is
it the tree that you think of when you look
at your calendar and oh gosh, I got a water
the trees? I have no idea plant a seed is
all I can do.

Speaker 1 (01:02:15):
What did you think of the movie Inception Love Inception.

Speaker 2 (01:02:19):
I think Inception totally rocks, And the biggest thing it
did was make the world think about their dreams more
and make them think dream incubation was possible. So when
people come into lab, I just say, oh, it's kind
of like that Inception thing. They say, oh, great, cool,
I've seen the movie. I know exactly what's about to happen.

Speaker 1 (01:02:33):
Last question, what does dream engineering look like in ten years?

Speaker 2 (01:02:37):
So dream engineering is what we've named this whole field.
So you know, doctor Harrison Chow, who's working on the
anesthesia dreams, is calling himself a dream engineer. But also
Michelle Carr, who's working on better versions of imagery rehearsal therapy,
she's calling herself a dream engineer. It encompasses a lot
of folks who are doing everything from chemistry to behavioral technice.

(01:03:00):
So my hope is that in the same way you
can go to a sleep clinic, you go to a
dream clinic. For a really large portion of insomniacs who
come into a sleep clinic, those people do not have
a sleep problem, they have a dream problem. They are
sleeping the right amount of hours, but the whole time
they're in end too, which about fifty percent of the
night they're having anxiety dreams. The dreams are I wish

(01:03:23):
I could fall asleep, I wish I could fall asleep.
I hope that I fall asleep. But they're literally asleep
that whole time, for hours as they wake up, and
it's called sleep state misperception. All the people coming into
an insomnia clinic saying hey, please fix me, and the
doctor says, you're actually sleeping seven hours, that's a dream problem.
I want dream clinics for those people. All the people
with PTSD who are coming in repetitive nightmare. What percentage

(01:03:45):
of them actually go to a clinician less than ten percent?

Speaker 3 (01:03:48):
Great?

Speaker 2 (01:03:48):
I want these tools, these dream entering tools, on every
phone in every bedroom of a person who's struggling. There's
these other areas we didn't even talk about. In substance
use disorder, for people who dream of a substance that
predicts relapse onto that substance. All those people on their phone,
they should have image rehearsal therapy to help them with
those dreams. Across insomnia, across addiction, across depression, across PTSD,

(01:04:11):
across pro drums for Alzheimer's and pro drums for Parkinston's
I hope the dream entering is a way to make
the world care and then to give them an easy tool,
as easy as the tools we have for meditation. Hey,
give me ten minutes of your time to do a
few techniques before bed, and we'll change that internal world
you have. Make it a little more joyous, make it
a little more vivid, make you live a little more

(01:04:33):
of those twenty six years.

Speaker 1 (01:04:40):
That was my conversation with Adam Harror Horowitz. In neuroscience,
we often think about the work that dreams are doing
in the background, like consolidating memories and exploring possibilities and
stitching together the threads of our lives. But at the
same time, there's the question of the experience that we're having.
Because even when we take a mechanistic view, like the

(01:05:02):
defensive activation theeriy where dreaming keeps the visual cortex active,
that still leaves open the question of what the experiences
mean for us, in other words, what they feel like
and what they do to us. And that's where Adams
work with lots of his colleagues and dream engineering, that's
where that work comes in because it's pointing to the

(01:05:23):
fact that dreams are part of our mental life, and
a part that largely gets ignored. We can live our
entire lives treating dreams as disposable things that evaporate in
the morning. But what happens if we shift that stance.
What happens if we decide that this internal world matters

(01:05:43):
once we engage with it, other possibilities open up. We
can begin to notice patterns that might not be obvious
during the day, like recurring themes, and we can begin
to reshape experiences like nightmares by introducing small changes. We
can begin to use dreams as a space for exploration,

(01:06:05):
for creativity, for problem solving, for reflection, and maybe we
can start to see dreams as continuous with waking thought,
part of the same process of the brain modeling itself
and its past and its possible futures. And the larger
horizon is that as dream engineering continues to develop, we're

(01:06:25):
going to increasingly find ourselves in a world where people
actively shape their inner experiences during sleep. We'll have dream
clinics and dream interventions and dream based diagnostics. I think
there will be a lot of questions here as we
move forward, like if you could guide your dreams how
much should you? Maybe unpredictability is part of what makes

(01:06:48):
streaming useful for allowing unexpected associations, So there will be
questions about what is gained and what is lost. But
the big picture is this. You spend a huge chunk
of your life in this state, years and years of experience,
worlds built and dissolved every night, conversations, fears, flights, reunions,

(01:07:13):
all of it unfolding in the dark. So given that
that much of your life is happening in a place
that you rarely remember, what might you discover when you
start paying attention. Go to egleman dot com slash podcast
for more information and to find further reading. Join the

(01:07:36):
weekly discussions on my substack, and check out and subscribe
to Inner Cosmos on YouTube for videos of each episode
and to leave comments until next time. I'm David Eagleman
and this is Inner Cosmos.