Ep44 "Why can't you tickle yourself?" (Time Traveling Part 2)

Why are people who can't remember their past also unable to picture their future? Why do we get so anxious about the world changing around us? What should you advise the president if we find ourselves at war with extraterrestrials? And what does this have to do with Wayne Gretzky, or the Greek goddess of memory, or hitting a bottle to get ketchup onto your French fries? Join this week's episode to find out about one of the most important things brains do: simulations of possible futures.

Episode Transcript

Speaker 1 (00:05):
Why can't you tickle yourself? Why don't hippopotamuses tell stories
around the campfire? What would I advise the president if
we find ourselves at war with extraterrestrials? And what does
any of this have to do with Wayne Gretzky or
the Greek Goddess of Memory and her children? Or poking

(00:28):
your finger into the side of your eyeball, or hitting
a bottle to get ketchup onto your French fries? And
why do we get so anxious about the world changing
around us? Welcome to Inner Cosmos with me David Eagleman.
I'm a neuroscientist and author at Stanford and in these

(00:48):
episodes we sail deeply into our three pound universe to
understand why and how our lives look the way they do.
Today's episode is about one of the most important things
that brains do, which is the simulation of possible futures.

(01:15):
The way we teach about brains in the classroom usually
has to do with the brain figuring out where it
is and what is happening around it, Like it detects
touch on its skin, and it detects photons from the
environment out there, and it picks up on sound waves
that are happening, and it stitches all of these together
in the massive hurricane of electrical spikes that race around

(01:38):
in the silence and darkness of your skull, and all
of this neural information allows you to put together a
picture of what is happening in the world out there,
and that is what allows you to operate in the
world and to catch that fish and put it in
your mouth, and to run from the predator, or more hosaically,

(02:00):
to find the right empty parking space, or tell the
cashier what you want from the fast food menu, or
apply the brakes on your bicycle when there's a pothole
in the road. So the brain gathers data from the
world around it so that it can operate inside of
that world. But what we talked about last week was

(02:21):
something surprising, which is that the brain doesn't spend all
of its time in the present. In fact, a lot
of its experiences are not in the here and now
at all, but instead in the past. Your brain holds
on to data about previous events in your life. In

(02:42):
other words, what happened and who was there, and what
the spatial configuration of the furniture was in the room,
and the building, and you spend quite a lot of
your time recalling that past. This whole process is what
we summarize as memory. And what I emphasized last week

(03:03):
is that we spend a good deal of our existence
unhooked from the here and now, and instead we time
travel to that past. But why do we care about
the past? This is for one reason only. We do
it to better simulate possible futures, and that's what today's

(03:24):
episode is about. It turns out we spend an enormous
amount of our time traveling to the future. Our time
travel is not one way, it goes both directions. We
simulate possibilities. We think of what our actions could lead to.
If I say this, then my spouse might say this
other thing back to me. If I open this cabinet

(03:47):
over here, I will expect to find soup cans in there.
If I do XYZ, I'll impress my boss, and I'll
hope to get that promotion. We walk down long paths
of possible chess moves that we can play in our lives. So,
as I said, we spend the vast majority of our
lives not in the here and now, but in the there,

(04:08):
and then in either direction. We spend most of our
days in daydreams and stories and confabulations in life narratives
of reminiscence and future projection. When you tally all this up,
the years that we spend in the realm of fantasy
outstrip the time that we spend in the present. So

(04:32):
why do we simulate the future? Well, first, it's much
more energy efficient to do that than to try everything
out in the real world. If I have to haul
this rock over there, I can sit and simulate several
possible approaches. I can imagine myself picking it up and

(04:55):
carrying it over there. But then I realized, now I'll
never be able to get it o over that big gulch. Well,
thank goodness that I ran that simulation from the comfort
of sitting on the ground and thinking about it. That's
super energy efficient, and I can try out different methods
of hauling the rock. What if I use a rope

(05:15):
or a wheelbarrow or a catapult. I can try out
all these different things without breaking a sweat or burning
any calories besides the few calories required for simulation, which
is orders of magnitude less than employing my muscles to

(05:36):
move my multi trillion cell body. Around in the world,
and this kind of simulation, this is what we humans
do all the time. So imagine you are a fire
chief and you and your team roll up on a
new fire that's engulfing a building. Your job is to
quickly make predictions about how to best position your team. So,

(06:00):
given your past experience with the world, you mentally simulate
different layouts and you evaluate their effectiveness, and then once
you've simulated a great plan or the best of what's available,
then you set it into action. In the real world,
you don't have to try out every single thing in

(06:21):
the physical world, and it's not just about energy efficiency.
More generally, the reason you simulate possible futures is because
it's much less dangerous than trying everything out in the
real world. So let's say you're parked in your car
and you need to go to some door, but there
is a dog barking at you, So you run a movie.

(06:44):
You simulate what would it be like if I make
a run for that door, and your brain might run
that and decide, you know what, that's not worth the risk.
So your brain simulates other plans, like maybe you stay
in the car and you dial the owner, or maybe
you crack the window and you yell for the owner.
Things like that, your brain doesn't actually have to run

(07:04):
the risk of confronting the dog. Or maybe some big
guy cuts in front of you when you're waiting in line,
and you might simulate all kinds of things that you
want to do to him in including calling him names
or pushing him in the back or whatever. But you
are usually better running the simulations in your head and

(07:24):
not taking advantage of all the things that you could do.
As the philosopher Carl Popper put it, simulation of the
future allows our hypotheses to die in our stead, So
intelligent brains do not want to engage in the expensive

(07:46):
and potentially fatal game of physically testing every action to
figure out what the consequences are. Instead, it is more
efficient and safer when possible to envisage consequences of a
proposed plan before you actually execute it. So by learning

(08:06):
up the rules of the world and simulating possible outcomes
and evaluating each one of them, your brain can play
out scenarios without the risk and expense of attempting them physically.
For this reason, prediction of possible futures is one of
the highest priorities of biological systems. Now, fascinatingly, there hasn't

(08:30):
been that much direct study of how brains do this,
mostly because of the difficulty of observing it in action,
because the whole purpose of mental simulation is to make
action unnecessary, and the traditional way that we study the
brain is to correlate something in the brain with an

(08:54):
action and explicit behavior. So this is what makes it
challenging to study simulation of the future. Nonetheless, what I'm
gonna tell you about today is the way that we
can pull together scattered data to begin to understand how
brains build possible worlds. So let's get started. The important

(09:15):
place to start is with an idea that I've talked
about a lot on this podcast, the idea of the
internal model. The idea is that the brain's job isn't
just about detecting and reacting in real time, but it's
about constructing a model on the inside about what's happening
in the outside world. It's like you're running a simulation

(09:38):
there in the silence and the darkness of the brain.
And the key is that the internal model can emulate
possible scenarios. Now, one of the first places that we
see this kind of simulation get studied is with our
physical interactions with the world. For example, think about when
you hold a ketchup bar bottle in your left hand

(10:01):
and you pound it with your right hand to try
to get the ketchup to come out onto your French fries. Now,
when you do that, your left arm doesn't move very
much when you pound with your right hand because your
muscles tighten up. But just the right moment. Now, if
you want to try an interesting experiment, just hold the
ketchup bottle and have a friend pound the bottle, and

(10:25):
what you'll see is that your arm moves a lot.
You can't keep your arm still when somebody else is
hitting the bottle, But when you hit it, you are
the one making the action, and your brain knows how
to simulate what's about to happen in this case, that
there's about to be a lot of pressure on your arm,
and so it can deal in real time with counterbalancing.

(10:49):
That This is the brain predicting something simple that hasn't
actually happened yet. Your hand is about to hit the
bottle and your arm tenses up in expense. Now, why
does the brain care about prediction? Well, if your brain
can simulate things into the future that can speed up

(11:10):
your response time. And this is really useful for potentially
dangerous things, like if you need to dodge a rock
that's being thrown at you, or if you need to
catch some prey and you can figure out where it's heading.
This reminds us, of course, of the great hockey player
Wayne Gretzky, who said, I skate to where the puck

(11:32):
is going to be, not to where it has been. Now,
you don't have to be Wayne Gretzky for your brain
to be predicting the next step of everything around you. Why,
because your brain's architecture is built to anticipate everything in advance.
How does it do this Well, First, your nervous system

(11:55):
doesn't just send out motor commands like move your arm.
It also sends out copies of that motor command along
other telegraph wires to let other parts of the brain
know that the command was just sent out. So this
is what's called an efference copy, and that has all

(12:16):
kinds of effects on what happens next. For example, when
you move your eyes around, your eyes are jumping about
three times per second. The world seems to remain stable
visually and This is because of the efference copy that
tells the rest of your brain, okay, visual cortex, get ready.

(12:38):
The whole world is about to streak past to the left.
So your visual cortex, which is locked in darkness, isn't
surprised when the eyes suddenly make a jump and the
data is now all different. Now contrast that with what
happens if you get a friend to gently push your
eyeball from the outside. Now, the visual world appears to shift.

(13:04):
In the first case, when you are moving your eyes voluntarily,
the eference copy tells the brain, hey, a move is
coming up, and that suppresses visual motion detection. But in
the second case, when somebody pushes your eye, the absence
of an effherence copy means, hey, that movement isn't mine,

(13:24):
it's external, And so you perceive visual motion in the world.
And I'll give you another example. What happens every time
you blink your eyes. When you do that, for about
a tenth of a second, the world goes dark. But
you don't perceive that because you know it is coming.
Your brain systems that send out the command to blink

(13:47):
the eyelids also let the visual system know, hey, this
is what's about to happen. This way, your visual cortex
can anticipate that's about to happen, so the blink gets ignored. Now,
if you don't believe me, maybe you think the blink
is just too fast or something like that. Just sit
in a room and have your friend flick the light

(14:08):
switch really fast so that everything goes dark for a
tenth of a second. You can't miss that. It's really
obvious since it wasn't you that caused the darkness, you
notice it. So, as we see with the ketchup bottle
and eye movements and blinks, brains make predictions about the
consequences of your own actions. And a great example of

(14:33):
this is tickling. It turns out that if someone is
coming after you and sticking their fingers under your underarms,
it makes you laugh uncontrollably. It tickles. But I don't
know if you've ever tried this. It turns out you're
not able to tickle yourself. Why not, It's because of

(14:54):
the simple fact that your own actions are predictable by
your brain. You can't surprise your left underarm with your
right hand because your brain is the one driving the
fingers of the right hand, and it determines exactly when
to move the fingers, and so there is zero surprise

(15:16):
when the left underarm senses that for tickling, you require unpredictability.
That's the whole trick to a tickle. When people study
this with brain imaging fMRI, they find that when you
try to tickle yourself, you get activity in the primary

(15:36):
somatosensory cortex, meaning your brain is detecting that there's a
feeling there. But the activity doesn't go further. It doesn't
activate all these other downstream areas that come online when
someone else is tickling you, like the secondary somatosensory cortex
and the anterior singular cortex. Your brain sees the tickle

(15:59):
coming and discounts it. In fact, other neuroimaging studies find
that these brain areas that come online when you're getting tickled,
these same areas become active when you are simply anticipating
a tickle. When somebody's fingers are moving close, your somatis
sensory areas start to go to town. Now it turns

(16:21):
out there is one way that you can tickle yourself,
and that is if you build a little device that
inserts randomness so you can no longer predict the tickle.
So imagine you set up a little machine where you
are moving a lever around like a joystick, and that
controls a feather that tickles your left underarm. But in

(16:46):
between the movement of the lever and the movement of
the feather, the computer injects random time delays. That way,
your brain can't know when the movement in your under
is going to occur. And now you rescue the tickle.
With the help of a little bit of technology, you

(17:07):
can tickle yourself. Importantly, it also turns out there is
one group of people who are able to tickle themselves,
and that is people who are suffering from schizophrenia. So
in episode thirty three, I talked about my hypothesis that
schizophrenia might be in part or in whole a disorder

(17:30):
of time perception. So in this light, it's very instructive
that people with schizophrenia can tickle themselves. This suggests that
they're unable to predict their own actions and how those
actions will lead to the next sensations. This inability to
understand one's own actions, this is a general deficit that

(17:53):
we see in schizophrenia. People will have a hard time
distinguishing things they call from things they didn't cause. A
person with schizophrenia will say something like, my hand picks
up the paper clip, but I'm not the one controlling
my hand. What my hand does has nothing to do
with me. And of course you've heard of things like

(18:15):
auditory hallucinations in schizophrenia. In that other episode, I talked
about how we all have an internal dialogue. You generate
a voice and you listen to it, and in schizophrenia,
the timing seems to be slightly off such that the
internal voice gets attributed to somebody else. Interestingly, my colleague

(18:37):
Chris Frith and his collaborators ran a study with people
who had schizophrenia, and they found that if a person
has schizophrenia but does not have auditory hallucinations, then they
are more ticklish when other people tickle them. But if
a person with schizophrenia does have auditory hallucinations, then they

(18:59):
can tickle them themselves. They judge no difference between someone
else tickling them and them tickling themselves. They are no
longer making the appropriate predictions. Okay, so this gives us

(19:29):
a sense of how our brains, under normal circumstances work
to constantly make predictions. And it's not just about predicting
things about your own actions and sensations, but more generally
about anything to do with the outside world. And the
key is that our brains are not simply reactive, but
they have these internal loops that are constantly making predictions

(19:54):
about what comes next. And having this kind of architecture
it allows brains to do not just stimulus response, but
instead to make predictions ahead of actual sensory input. So
think about trying to catch a baseball that someone is
tossing to you. If your brain was just doing feed

(20:16):
forward analysis of these signals, you couldn't catch the ball.
There'd be a delay of hundreds of milliseconds from the
time that the light strikes your eye until you could
execute the motor command of putting your hand up, and
your hand would always be reaching for a place where
the ball used to be. We are able to catch

(20:37):
baseballs because we have deeply hardwired internal models of physics,
and these internal models generate expectations about when and where
the ball is going to hit. It's making predictions about
the future. Now, how does this sort of prediction play
out in your daily life? Because it's not just about

(20:59):
hitting catch a bottles and catching baseballs. But prediction applies
to every decision you make about what you're going to do.
Let's say you're trying to figure out what you need
to do in an hour from now. You have a
bunch of things on your to do list, but given
the constraints of space and time, you can't do everything
at once, and so life is a constant series of choices.

(21:23):
So let's say you're looking at these choices. One you
have to write a very long and specific email for
your boss. Two you're thinking of going downtown to get
a boba tee, or three you're considering going to the gym,
which you've been promising yourself that you're going to do
for some days now. So how does the choice actually
get made in the brain. As far as our conscious

(21:46):
minds go, we get very little access to the details.
It just seems like, oh, okay, I've decided to do
this instead of that, but you don't necessarily know why.
But the last several decades of neuroscience have surfaced how
this actually happenins We run the simulations and we feel them.
So when you think about writing that email, your brain

(22:09):
actually runs the little movie of you doing that, and
possibly the ache that you might feel in your shoulder
from typing too much, and also the satisfaction at finishing
that task. Then your brain runs the simulation of going
and getting the boba t how delicious that will be
and how satisfying it'll be. And you also run the

(22:32):
simulation of going to the gym. It's gonna be a
little costly for you in terms of time, and it
might make your muscles hurt, but boy, are you gonna
feel great when you're done. You'll feel so satisfied. Now,
what happens is your brain runs all these simulations and
you feel the emotions with each one. Now again, this

(22:53):
happens essentially entirely under the hood. For most of the
decisions you make in life. You have no conscious access
to how you made the final call, But your brain
is simulating the possibilities, and you experience each one with
your emotions and often with physical sensations. And this is

(23:13):
how we weigh choices against one another. This is how
we determine our paths in life. You feel the pain
or the pleasure from your predicted futures. You think of
yourself in future times, and you get to live out
those little movies. Now like everything in our brains. We

(23:35):
think this just automatically works, but in fact we have
very particular networks that need to be in place and
working well for this to function. And the reason we
know this is because some people get damage to a
part in the front of their brain, the venture medial
prefrontal cortex, and they end up displaying a very strange

(23:57):
and unexpected symptom, which is, if you give them some
choice to make, like which restaurant do you want to
go to tonight, they can articulate everything about the decision,
but they can't decide. They can't land on a decision.
So what's going on here? Well, patients like this have
been studied by neurologists like Antonio Demacio and his colleagues,

(24:21):
and what's happening is that their brain can run a
quick future simulation, but it has become disconnected from the emotions.
So the different future simulations all feel the same way.
They all feel neutral, and therefore there's no way to
distinguish any choice from any other. In other words, you

(24:43):
need to feel the outcome of a simulation in order
to do decision making. So it turns out that under
normal circumstances, there's a core network of brain areas that
are involved in prospect, which just means seeing ahead. So
I want to give you a very quick sense of this.

(25:05):
So you've got this area of your brain, the venture
medial prefrontal cortex, which connects to areas involved in emotion
like the anterior insula and the amignla, and it also
connects to lots of other areas in the brain. And
activity in this area essentially specifies the things that are
pertinent to your current needs and goals, and this is

(25:27):
what guides the construction of relevant scenarios. Now, there are
a number of other brain areas that show up in
this core network. One is called the precuneus, and this
maps the locations of things in space, so its involvement
contributes to a spatial context for imagine scenarios and the

(25:50):
features inside of that. There's another area called the temporopridal junction,
and you see that area become active during the detection
of targets or events around you that are relevant to
what you're trying to do at this moment. This area
seems to run and do the same thing even in

(26:11):
your imagined scenarios. And then you've got an area called
the superior temporal sulcus, which is involved with lots of things,
but one of them is about interpreting social cues. So
one idea is that this area helps to specify other
individuals in their actions within imagined scenes. And then you

(26:34):
have the hippocampus. And one thing that's known is that
when you get damage to the hippocampus, that seems to
mess up all of the spatial coherence of a recalled
or imagined scene. So patients with hippocampal damage they can't
picture a specific place or detailed surrounding events. Here's one

(26:58):
way to think about this.

Speaker 2 (27:00):
When you move in a virtual reality world, the goggles
keep track of where you are and all the objects
and how things change when you move.

Speaker 1 (27:12):
And this is similar to what's going on in the brain.
You have special cells in the hippocampus called place cells,
which help to translate everything into a framework where you
are at the center of it. And the idea is
that these cells are critical to your imagination of scenes.
So as you virtually move through your imagined scene, the

(27:37):
hippocampal play cells keep the scene coherent and consistent, just
like the VR goggles would. So this all suggests that
the hippocampus is crucial for tying together the activity of
other brain areas to construct this rich and coherent imaginary experience.

(28:00):
You have this brainwide network of areas that are involved
when you are imagining future scenarios, and all of this
is what helps you to experience the movie and to
feel the emotions. If you are just a robot who
rolled into a room, you would just sit there because

(28:21):
you wouldn't have any particular reason to prioritize writing the
email versus getting the Boba tee versus going to the gym.
You would have no way to weigh these against one another.
But we assign feeling to all of our future scenarios.
So the brain makes predictions. But how does it know

(28:43):
how to improve these through time. Well, imagine a fire
department in a city, and every time a fire occurs
in the city, they go wailing out of the station
to take care of it, and they suspect that a
lot of the fires are going to happen around the
warehouse district, so they put their trucks there so they

(29:04):
can take care of things quickly. But eventually they realize
that their prediction was wrong. It's actually another part of
the city that keeps catching fire. So the area at
the foot of the mountains where the trees are dense
and interwoven with power lines, So the fire department starts

(29:24):
putting their resources there where they need to be in advance,
and that reduces the energy they need to expend every
time there's a new fire because they're no longer being
reactive to fires at the foot of the mountain, but
now they're making good predictions about it. So cities do
this kind of thing, by the way, in terms of fires,

(29:44):
in terms of where they expect crime is going to occur,
in terms of where and when the power usage is
going to happen. Everything. So the key about this example
is that the fire department's first predictions weren't so great,
and the data tells them, oh, that could be a
lot better. It tells them that something could be adjusted.

(30:05):
And that is the same thing that happens in the
brain all the time. The brain tries to predict everything,
and it pays attention to what's called the prediction error,
which means the difference between what it thought would happen
and what actually happened. And you see various cells in

(30:25):
the brain, for example, in the dopamine system that are
responding not to the reward or punishment, but the prediction error.
In other words, how different the reward or punishment was
from what was expected. And these dopamine systems broadcast their
signals all across the territory of the brain to announce

(30:49):
that the prediction wasn't quite right, there was a prediction error,
and therefore something needs to be adjusted. So our brain
has the architecture to make predictions and adjust them all
the time. And what is all this architecture of the
brain tell us. It tells us that the brain craves predictability.

(31:12):
Now why does it care about predictability? Well, first of
all because of energy efficiency. Because if you can predict
that something is going to happen, then you don't have
to burn up all this neural energy on it. You
already know it's coming. But if something is a surprise,
the brain turns its vast attentional mechanisms to it and

(31:34):
has to burn a lot of calories on understanding what
the heck just happened and eventually reshaping the internal model
to account for that. In the future, all of this
would be fine. Maybe if we could plug ourselves into
a wall, but instead we are mobile creatures who run
on batteries. We have to constantly find food sources and

(31:56):
stick them in our mouth so that we can have
enough energy to get to the next food source. So
mother nature evolved us to be highly efficient creatures. And
what we do is we make ourselves massively efficient by
predicting away the future. And this is, by the way,

(32:16):
why the method of torture referred to as the Chinese
water torture is so aversive to us. The idea is
that a drop of cold water drips onto your head,
and then let's say five seconds later, the next draw hits,
and then the next one three seconds later, and then
the next drop eight seconds slater, and the next one

(32:39):
six seconds slater, and then four seconds and suddenly one second,
and you get the idea. It's unpredictable. Your brain is
constantly trying to say when an event is going to happen,
and it's constantly having to pay attention here because it
can't make a good prediction. And perhaps you've never experienced
that form of torture explicitly that most of us have

(33:03):
at some point in our lives, had a leaky faucet
at our house, and this can often be just as
bad if it never falls into a rhythm, it goes drip, drip, drip.
We love rhythm because we can predict it away, and
anything that is unpredictable continues to demand all our attention. Now,

(33:28):
I'll just mention that my colleagues and I have proposed
in various places that maybe the brain activity that we see,
the spikes and neurons, these represent just the part of
the world that is unpredicted. In other words, silence is golden,
and the brain spends most of its efforts trying to

(33:50):
make perfect predictions of the world and burn that down
into the circuitry of the brain so it doesn't have
to use any activity. Now, of course, of course, the
world is way too complicated to ever reach perfect prediction.
Everything changes all the time, and so the speculation is
that the activity that we can measure in the brain,

(34:12):
whether that's with electrodes or fMRI or whatever, really that
activity just represents the surprise, the thing that the brain
didn't see coming. In other words, if you show a
yellow ball to a monkey and you find cells in
his brain, say in the visual cortex that respond vigorously

(34:32):
to that visual thing, then we say those cells prefer
yellow balls, or more colloquially, we say it likes the
yellow ball. But could it just be that the appearance
of a yellow ball was unpredicted by the system and
the cells activity is merely a reflection of that. This

(34:53):
is consistent with the fact that if you hide the
ball and then reshow it to the monkey five seconds later,
and then you do that again and again, the response diminishes.
This is known as repetition suppression, and it's not merely
about fatigue of the cell. Instead, it's the fact that
the monkey's brain knows that you're about to show the

(35:15):
stupid ball again, and so it has a prediction of
what is coming. And when it knows what is coming,
it doesn't have to burn any energy. We'll come back
to this in terms of our deep desired to have
predictions about our lives in just a few moments. But
first I want to ask can species other than human

(35:36):
beings engage in prospection and imagination? This question is difficult
to answer because animals can't verbally report their experiences to us.
Some researchers think, well, maybe animals lack the capacity for
this sort of thing, but in fact they do share
much of the same circuitry that we've been talking about,

(35:59):
and careful observation their behavior suggests that they have some
features of episodic memory and prospection. For example, look at
the scrubjay, which is a bird that stores food away.
It can recall not just where it hit a particular item,
but also what that item was and when it was stored. Now,

(36:23):
skeptics say, okay, look, maybe this is just procedural memory.
It's like a basic algorithm that's running. It's not conscious,
and it's just driven by the needs of the moment.
But the scrub jays also appear to cash food in
a way that reflects anticipated future needs. It's not just
their current motivational state. And when you look at rats,

(36:48):
when people do direct recordings of the hippocampus, that suggests
that they too might engage in prospection and recollection. So
as a rat move through adjacent places, you have these
hippocampal play cells that fire off in sequences, and these
play cells sometimes replay the activity sequence when the rat

(37:13):
is not moving, sometimes even when the rat is sleeping
after the experiment is over, and these cells can also
pre play a sequence of activity before the rat has
started to move along the route. So, for example, if
the rat has to go down a hallway and then
turn right or left, this is called a te maze,

(37:35):
and it's trying to decide which path to take. You
can see these play cells pre play one route and
then the other, as if in consideration of both these scenarios. Now,
this research is still early, but these kinds of findings
might increasingly point us in the direction of at least

(37:57):
roughly gauging whether our animal cuts and have internal experiences
of time travel the way that we do. So what

(38:23):
I've told you so far is how the brain predicts.
But how does it know how to do this? How
does it make good predictions about the world. So suppose
you're hungry and you decide you're going to get something
to eat. Where should you go? You need to have
a map in your head of all the nearby choices.

(38:44):
Then you have to decide which one would most likely
satisfy your current craving. And so to plan your excursion
you need to go through your past experiences of meals
and the places on your map. So you've got that
inexpensive tie restaurant which is the closest, but the food

(39:05):
there you remember, was too spicy for your taste. And
the food truck over there they make great burritos, but
it always has a really long line in your past experience.
And the fast food joint over here has fries that
are a little greasy, but you'd rather take the grease
than the spice or the long lines. So, putting together

(39:26):
the experiences of your recalled past and your imagined future,
you decide on your option. But the past and the
future are intertwined in most of our decisions. In other words,
the thing that allows the brain to construct possible futures
is memory. Memory is what allows us to write down

(39:50):
information and then use that as building blocks to build
out our future scenarios. Now, interestingly, that's not even a
new idea. Aristotle suspected this, as did Galen and all
their medieval commentators. They all emphasized memory as the key
tool in making successful predictions for the future. And in fact,

(40:14):
something that I find very interesting, presumably coincidental but maybe not,
is that in Greek mythology, the goddess of memory, Nemazine,
is the mother of the nine muses, who are the
goddesses who spark the imagination. In other words, memory is
the mother of imagination. Now, my friend and colleague Jeff

(40:38):
Hawkins made the argument that what we call intelligence boils
down to the brain's ability to make good predictions about
the world based on stored memories. In his version of
the memory prediction paradigm, the cortex is a pattern recognition
machine that breaks complicated events into smaller bunks. It stores

(41:01):
experiences in a way that reflects the structure of the world,
and then it's springboards off these known experiences to make predictions.
As Hawkins puts it, intelligence is the capacity of the
brain to predict the future by analogy to the past. Now, fascinatingly,
in two thousand and seven, Demisesabis and his colleagues at

(41:24):
London's Institute of Neurology made this really striking observation that
patience who had damage to their hippocampus not only had
amnesia for past experiences, but also couldn't imagine new ones.
So if you ask a patient with this brain damage

(41:45):
to remember his past, he can't do that, and we
expected that. But now you ask him to imagine the future,
and he just can't do it. You ask him to
imagine standing in a museum full of exhibits and he'll
say something like there's not a lot coming, I'm not
picturing anything. Or you might say, hey, look, imagine going

(42:06):
on a vacation to the beach. Really picture yourself lying
there on the beach, and describe the scene to me.
And the person might be able to say, well, there's
a blue sky, or maybe they describe an isolated sound,
but that's it. Otherwise they just draw a blank. And
by the way, if you provide the person with pictures

(42:28):
and sounds and smells to help them along, that doesn't
help them to imagine the scene. So unlike a healthy control,
the patient with hippocampal damage just can't simulate any vivid details.
It's not like an episode to them the way that
your imagination is. Their descriptions, if they exist at all,

(42:49):
are very unspecific. So the healthy control can describe a
spatial layout and people being present, and descriptions of the
smells and sights and sounds, thoughts and emotions they might have,
and actions they might take. But the patient with the
hippocampal damage can't do any of that. At best. What

(43:10):
they're able to come up with has a lack of
spatial coherence. The imagined experience is just a collection of
fragmentary sensations instead of a unified episode. In a particular setting,
you can ask them what they'll see if they go
over to a shopping mall, or what they might want
to eat if they go to a restaurant, and they

(43:32):
just draw a blank. They can only put together a
few details that are not well connected. So the deficits
in their memories apply to their simulations of the future
as well. Now, how do we understand this in terms
of the circuitry. Well, the key came from brain imaging
studies in the past two decades, which have uncovered that

(43:55):
there's a network of regions involved both for remembering the
past and imagining new ones. It's the same areas, so
the hippocampus and its surrounding area. That's one part of this,
but we also find several other areas like the medial,
prial cortex and prefront areas, and the lateral, temporal and

(44:15):
lateral pride lobes. All these regions are important for elaborating
on the details of imagined and also for remembered episodes. So,
in other words, the brain's episodic memory systems, which we
discussed in the last episode, are just as important for
imagining future experiences. In other words, this core network underlies

(44:41):
mental time travel in either direction. So this leads to
a really interesting thought about something, which is that if
recall of the past and simulation of the future both
use the same network, then maybe what we mean by
memory is something more like simulation. So I want to

(45:02):
propose this hypothesis that memories are not the fundamental thing,
but instead simulation is the fundamental thing, and memories are
just a special type of simulation. A memory is merely
a simulation that's pinned down to always flow a particular way.
And if this is the right way to look at it,

(45:23):
maybe what we call remembrance we will someday call resimulation.
So instead of dividing the territory into memory and prediction,
they may in fact be one thing. Any context is
run through a simulation to predict the outcome. So brains

(45:44):
simulate possible futures, and we constantly function by making predictions
about everything in our lives and our communities and our
nation and the world. But I want to be clear
that even though we use the word prediction, there's no
guarantee of accuracy. We are actually pretty bad at capturing
the future. As the baseball catcher Yogi Bearra said, it's

(46:07):
tough to make predictions, especially about the future. Why is
it tough. It's because we only simulate based on our
experience in the world. So if you've never seen something before,
you're going to have a pretty bad prediction about it.

(46:28):
For example, futurists make all kinds of predictions about the
next decade or two, and most of them turn out
to be wrong. In one study of famous forecasters, it
was found that their predictions were between ten to fifty
percent accuracy, which certainly isn't that great. But it's not
just futurists. It's all of us, with most of our

(46:49):
predictions about our lives, and by the way, our inability
to see the future, well, this is why we have
the existence of magicians or mystery writers or scam artists.
These are people who take advantage of the fact that
our ability to predict the future is not very good.

(47:10):
The magician knows that we're going to predict the location
of the object incorrectly and then will be surprised. The
mystery novelist knows that he can lead us down a
garden path and that we will extrapolate incorrectly in the
direction that he wants us to, so we don't correctly
see what's going to happen. The scam artist does the

(47:34):
same thing, but in real life, getting our brains to
see a vision of success that doesn't actually match with
what's going to happen. Now, I'll just note something here,
which is that's sometimes our inability to make good predictions
that helps us. So take the origin of the Oxford
English Dictionary, where a professor who loved words said, you

(47:57):
know what, I'm going to write down a definition for
every word there is. This can't take very long, especially
if I recruit help from others, which he did. But
despite an insane amount of work, the Oxford English Dictionary
finally got finished eighty years after his death. He would
have never started it had he been a good predictor.

(48:20):
So there's something useful about our optimism bias in the
form of our bad predictions. Now, we've all experienced this
kind of bad prediction on smaller levels, where we assume
that some task is going to take us less time
than it actually does. We also experience this on the
level of most of our life trajectories, where we think,

(48:43):
all right, I generally know where my life is going,
but if you look back at any decade of your life,
you'll realize that your predictions generally weren't so good. Why
it's because the only way we can make predictions is
by leveraging our memories what has all ready happen to us.
The memories serve as building blocks, and all we're able

(49:06):
to do is use those building blocks to make versions
of the future, which is really just an edifice constructed
of the bricks that we've seen before. And that's why
we are so inherently limited in seeing what's coming. And
there's a very specific way that we're terrible at predicting
the future. We generally assume the future will just be

(49:29):
a straightforward extension of the present in our lives. We
assume that we have changed up to this point, but
we're going to remain about like this from here on out.
For example, when people think back to their childhoods, they
see lots of change in their own bodies and personalities
and beliefs, and also in the technology that surrounds them.

(49:52):
But when people are asked to think about the future,
they generally assume everything is going to be roughly this
same as it is now. Maybe you'll have a little
more gray hair, maybe your electric car will have a
longer range, but that's about it. We feel like we've
arrived after a steep path and now the world will

(50:13):
mostly stay fixed as it is. So we think things
are going to stay as they are. And nowhere is
this more true than with our predictions about technology. The
fact is, the world is changing faster than ever as
a result of the law of accelerating returns, which simply
says that the more technology advances, the faster the next

(50:37):
generation of technology is going to advance. And we find
ourselves now at the cusp of such fundamental revolutions, not
only artificial intelligence, but also nanotech and biotech and quantum
computing and room temperature superconductivity and energy storage and genetic
engineering and on and on. All of these things going

(51:00):
to weave together in ways that we can't currently imagine.
And we are standing on an exponential curve that's about
to rise at a steeper slope than we've ever seen,
but we can't see it clearly coming. Why again, it's
because we rely on our memories to paint our vision

(51:20):
of the future, so our predictions are limited to remixes
of our past, which makes it really difficult to anticipate
the significant disruptions heading our way. And there's one other
issue here for our lives, because we're always trying to
make good predictions and therefore save brain energy. We really

(51:43):
hate change. I mentioned before how we don't like the
dripping faucet because it's unpredictable, But this hatred of the
unpredictable applies to everything, including being told that things will
change in the future, like climate change. Climate change makes
people very anxious because you look at a map of

(52:04):
the world and you see that over the next x
number of decades people will be shifting around as temperatures increase.
And we hate that because fundamentally we feel most comfortable
if things stay exactly the way they are. Like you
want to imagine that your house, which is exactly two
hundred and fifty seven feet from the shore of the ocean,

(52:26):
will remain precisely where it is centuries from now, But
of course it won't, even if you put aside everything
about man made pollution. The shorelines always change. Where I
live in California, the beach used to extend miles farther out.
I'm talking about fourteen thousand years ago, and when the

(52:47):
ice Age ended and the glaciers melted, the sea level
rose and ate up all of that beach. And that's
why we don't find coastal settlements from the first people
from Asia who came across cross the bearing Land Bridge
and settled here fourteen thousand years ago. Things were totally
different at that time. For example, there was no San

(53:08):
Francisco Bay. There was no water there that was all
locked up in glaciers. When any of us who live
here look at the San Francisco Bay, we imagine it's permanent,
but of course it's not. If you were one of
the first settlers in North America, the place would have
looked very different to you. You could have walked across
from what is now the city of San Francisco over

(53:31):
to Alcatraz without getting a single drop of water on
your feet. It's easy to study geography retrospectively and say, wow,
that's interesting. But when we look in the forward direction,
we get very anxious at the thought that populations of
people will move around and borders will change. Now that's
not to minimize what's happening with climate change, but it

(53:54):
is to say that change has always happened. I mean,
the last little ice Age just ended in eighteen fifty,
where for five hundred years it was two degrees colder
in Europe and mountain glaciers expanded and people had to
move around. The only issue I'm pointing to here is
that even though the world has always changed, we want

(54:18):
it to stay stable now. We fundamentally want to imagine
the future of the world looking exactly as we know
it now. And you can see this as easily at
small scales as you do on the large scales. For example,
in this past month, there have been a new round
of company layoffs in Silicon Valley, and this makes people

(54:40):
so nervous and anxious. Now, most people who have lost
a job end up saying later that they're happy because
it opens up new opportunities for them and exposes them
to things they didn't even know. They didn't know and
they realized there was more out there to be experienced
in the world. And yet the change itself proves very

(55:02):
hard for people in the moment. It's as though their
brains are screaming out for everything to stay exactly the
same as it was. Why, Because we are creatures who
try to predict Our brains are designed to do that
to save energy, and the most anxiety producing thing for
the accuracy of our predictions is when the world itself

(55:23):
changes out from under you. As an example, I've been
on the boards of many organizations. When somebody resigns and
so much trauma for the board, there's a long discussion
about how to keep the organization together. Everybody's feelings are bubbling,
and then the conversation eventually turns to how this presents
a real opportunity for us to mix things up, to

(55:46):
inject new blood, to do things in a way that's
no longer stale. It's fascinating to watch the conversation always
follow the same trajectory, as though everyone is reading from
a script. It reminds me of a notion from The
Simpsons where Homer is anxious about something and Lisa, his daughter, says,

(56:06):
look on the bright side, Dad, Did you know that
the Chinese use the same word for crisis as they
do for opportunity, and Homer says, yes, chrisis as tunity.
The point is that change of any sort presents a
crisis to the predictive systems of the brain, but it
is eventually seen as an opportunity. The bottom line is

(56:30):
that we get used to the world and we don't
want things to change. So, given that our predictive ability
is not so great and that we don't want things
to change, how did we ever become so successful as
a species. Well, the first answer is science. When it
comes to predicting big issues in the real world, our

(56:50):
intuitions just aren't up for the task. Our brains always
make predictions, but human brains are small and they're not
nearly as good as groups of brains working together, and
the scientific method simply gives us away a set of
rules for working together to find the most accurate predictions. Fundamentally,

(57:11):
that's all science is figuring out the rules so we
can best predict the future. But I want to highlight
what I propose is a second reason that humans got
really good at predicting the future, and this is perhaps
a more surprising reason why our species has become a
runaway species, and that reason is storytelling. So literature like stories, novels,

(57:40):
and movies. This is critically important for the success of
our species because we can take one person's imagined stories
something they've worked out all the pieces and parts of
over a long time, and that author can make that
scenario real for us, He can reify it. So stories
allow us to experience possible futures. Just think, for example,

(58:05):
of the nineteen eighties movie The Day After. It was
about nuclear war and what it is to wake up
the day after America has been turned to rubble by
nuclear bombs. It took something that required an unusually rich
imagination and it allowed us to see it, to experience
a situation that otherwise would have remained purely conceptual. And

(58:29):
this is why stories are so important. They allow us
to live in worlds that we otherwise would not, and
then that gives us new memories that we can use
as new building blocks to see a little farther out
than we would have otherwise. In this sense, literature allows
us to get out of our heads and share the

(58:51):
creative headspace of someone else who has thought down a
particular path, probably in great detail, and then we get
to enjoy that person's guidance. And the key, as far
as we can tell, is that other species, for example,
hippopotamuses don't tell stories around the campfire. And it's not

(59:11):
just hippopotamuses, but every other one of the millions of
species of animals on this earth. We have no evidence
that any of them tells stories. So the way I
think about this is that they just have a lot
less practice expanding beyond their own limited experience of the world.

(59:31):
But we spend a ton of our time imagining what's
not there. We are mental time travelers, and we use
other people's stories and books to get there. The class
that I'm teaching at Stanford this quarter is Literature and
the Brain. What I find so extraordinary about the active
reading is that we use a string of symbols to

(59:53):
fire up this whole imagination engine and to have us
live through scenarios. And I propose that we have come
to beat out every animal species on the planet, including
lions and tigers and bears, all of whom could tear
us to shreds easily. We have beat them out because
of stories. We have these fierce animals in our zoos

(01:00:16):
in every city, and they have no humans in their zoos.
It's not just about guns and spears, it's about planning.
We can capture them because we can outthink them. So
I posed at the beginning of the episode a question,
what would I advise the president if we found ourselves

(01:00:37):
at war with extraterrestrials. Well, here's what If we land
on a planet with fierce bug like creatures, we shouldn't
worry too much about their capacity to be anything but reactive.
We will probably be able to trick them, to outflank them,
to outthink them. But if we discover that these creatures

(01:00:58):
also have life libraries, we should quietly turn around and
sneak away, because it means they have exposed themselves to
thousands of other worlds beyond what they could otherwise experience,
and that cognitive practice makes them potentially a very wily opponent.

(01:01:21):
The degree to which an alien species has literature will
tell us how good they are at predicting possible futures
and developing rich scenarios of what ifs. It allows them
to expand their experiences far beyond a single head. So
let's wrap up what we saw today is that brains

(01:01:44):
simulate possible futures, and brains do this by relying on
the lessons of the past. This makes your memory the
mother of your imagination, just like the goddess of memory
gave birth to the muses. Now, in the last two episodes,
we've talked about running simulations in the backward direction, which

(01:02:06):
we call memory, and running them in the forward direction,
which is how we envision possible futures. But everything I've
told you so far is just a setup, because that
is just the beginning. Come back next week to see
how now we can leverage this concept of time travel
to deeply understand why our mental lives are as nuanced

(01:02:29):
and colorful and complex as they are simulating. Next week,
I'm David Eagleman, and this is Inner Cosmos. In the meantime,
go to eagleman dot com slash podcast for more information
and to find further reading. Send me an email at

(01:02:50):
podcasts at eagleman dot com with questions or discussion, and
I'll be making episodes in which I address those until
next time. Thank you for joining me in the outermost
reaches of the Inner Cosmos