This episode is brought to you by Brilliant. There is a school of thought that says our
bodies are little more than a vehicle and life support system for our brains. If that’s true, we might be able to transplant
the brain into some much cooler vehicles. One of the more popular notions in science
fiction is the idea that you might separate a person’s head from their body and keep
it alive, or even just their brain, and today we’ll be looking at this idea, how it differs
in form, function, and implications from a digital copy of a mind - mind uploading – and
what reasons and motivations we’d have for doing this, such as life extension, extreme
emergency medicine, resource scarcity, or even punishment. This is going to be one of our more food for
thought, fun episodes but as we’ll see today, it’s not likely to be something we only
see in sci-fi, it’s actually quite plausible we will see this option used in the future,
especially if digitally copying minds turns out to be harder than we expect or even practically
impossible. Speaking of food for thought, you might want
to grab a drink and a snack, and hit the like button while you’re at it. The idea of brains in jars has fallen mostly
into the realms of more humorous science fiction, like Futurama, in favor of the notion of mind
uploading which tends to seem the better option – and we’ll discuss why in a bit, but
it’s worth remembering these are very different technologies and in truth we’ll probably
be able to keep a brain alive in a tank a lot sooner before we could emulate a brain
on a computer. The other thing about mind uploading is that
it’s not just trying to emulate human neurons and tie them all together to replicate a human
mind, it’s trying to exactly copy a specific person’s brain, which is going to be a good
deal trickier, as you’re not trying to achieve a good likeness but rather an exact one. This is where we have tended to let fiction
and notions about technology run a bit away with us. There are many reasons you might copy a human
mind, but the big one is usually to make an immortal version of a person that you can
digitally back up. If the copy is right, then that person is
preserved but the original has simply been copied, and the original is still sitting
there in the brain scanner afterward unless for some reason the scanning process was inherently
destructive or lethal. Which is a nice handwave in fiction to remove
the original person from consideration but not a very likely one, that the scanning method
would just coincidentally need to vaporize the original. All the more so since we’re not likely to
ever be encouraging research on brain scanning that tended to vaporize brains. You have intermediate steps of technologies
and I’m not really sure how a partial brain scanner that vaporized neurons during the
process would be seen as useful to encourage further research and funding. So even if a destructive scanning method just
happened to be a bit easier than a non-destructive one it would probably lag behind in research
efforts to any non-destructive scanning approaches. That’s assuming we continue to practice
mostly ethical medical science, in ways that respect basic human rights... Of course you could also experiment on non-human
subjects or volunteers who donated their bodies to science, and that is an example of where
such a destructive scanner avoids the problem of a copying or killing the original brain,
as the person is already dead or dying. But even that has a problem because if you
start telling folks you want to vaporize their brain to research how to digitally resurrect
other folks most would probably say “Hey, wait a minute, why don’t I just get my brain
frozen instead, till you perfect the technique and can copy me?” And of course we can already do this, albeit
with extreme tissue damage at the moment. But in terms of cost, it’s not much more
expensive to freeze someone’s head than usual funeral costs or many common life-saving
procedures and mostly just because we don’t do it much and a lot of the money goes into
researching better preservation and resurrection techniques. In terms of raw cost, there’s not much difference
between a casket and a cryonic cask or dewar, especially one only big enough for a head. It’s certainly not a space and maintenance
issue either, cemeteries take up a lot of space and aren’t cheap to maintain, whereas
you could stick a hundred frozen heads or brains into a single cubic meter. As to keeping them frozen, liquid nitrogen
costs approximately as much per volume as milk, and a decent dewar only needs refilling
a few times a year – less if you had tons of them in storage together, and so preserving
a brain might be less than a dollar a year in terms of the nitrogen. This isn’t an episode on cryonics or cryogenics,
but it needs to be kept in mind because if we do perfect ways of freezing a brain that
are solidly non-destructive, which we can’t with current methods, that leaves the door
open to scanning or thawing that brain out one day with better technology and you’d
presumably see a very big uptick in folks choosing this. If you’re curious we could probably store
dozens of generations of brains on ice for centuries without it being any sort of significant
drain on the economy especially if it were done so frequently that it could create an
economy of scale for brain removal and tank production and maintenance. We don’t know which we’d be able to do
first, thaw a brain and stick it in a clone or android body or scan a brain, but even
if we could do both a lot of folks would presumably prefer their original brain over a digital
one. We often used to consider keeping the brain
in a jar a method of life extension once the body fell apart, but brains age too. A digital copy can be maintained indefinitely
by having backups and using existing data integrity methods. However we should keep in mind that if you
actually have the sorts of technologies needed for brain scanning – which is as likely
to be billions of nanobots running around your brain imaging each neuron as some super-MRI
scanning it externally – that tends to imply you can repair or build neurons too. Those of us already acquainted with the whole
copying issue often suggest a gradual neuron by neuron copy and replace with artificial
neurons, little machines emulating an individual neuron and replacing an existing one, but
you could presumably just set them to constantly repair and replace with regular neurons just
as easily. It might require more maintenance but it’s
the same maintenance your brain already does unaided anyway and neurons have much longer
lifetimes than most cells, though it depends on which type of neuron is under discussion
and we have multiple types. Needless to say brain cells can be damaged
and decay too, as any number of mental conditions associated with aging show us. Indeed that’s a lot of why the brain-in-a-jar
is an attractive idea, brain cells might not be able to live indefinitely, but they at
least have a lot more longevity than the body as a whole does. So if we never get practical medical nanotechnology
able to repair every single cell, the brain in the jar method might be a life extension
method we’d use. Also if you can copy a frozen brain onto a
hard drive there’s no particular reason to think it would be much harder to print
a new brain either. The level of intricacy and technology ought
to be parallel so scanning a brain to duplicate it on a newly printed organic brain as opposed
to a hard drive ought to be viable for those who’d prefer that. Neurons vary wildly in size and type but the
average in your brain would be around ten nanograms, which isn’t much but does mean
they’re each composed of trillions of atoms so we probably could grow various types of
generic grown neurons and use them in an advanced 3D printer a long time before we could be
doing precision atomic printing. It’s also a reminder that the brain isn’t
really atomic scale, though that is debatable in regard to the information storage on it. Still we probably don’t need to be duplicating
with atomic precision to properly replicate the same neuron for all practical purposes
like memory and of course they are constantly having new atoms added and old ones removed
too, those neurons need life support same as any other cell and are not identical to
what they were yesterday or ten years from now even if the neuron itself has kept going. Key notion is that there’s a lot of technical
difficulties involved in preserving or copying our brains, be it organically or digitally
and we have no idea which would be easiest or most effective, but even if one method
is easier it doesn’t mean it’s the one folks would prefer, so if you have multiple
methods that are doable and aren’t wildly different in effective costs and upkeep, folks
would presumably pick the one they liked. But a brain in a jar is not necessarily just
your brain as-is, placed in a jar. Once we’ve got that brain out of the skull
it could potentially be modified or grown larger. Which is another point too, a lot of the damage
we get or life support issues for the brain we have revolve around things like radiation
and contamination, and if you’ve got a brain in some big radiation shielded vessel with
a strict control of the nutrients going in, like no potassium radioisotopes just the non-radioactive
kind, or very refined chemicals to remove toxins and impurities, you probably can keep
that brain running a lot longer, and while it is changing with time, both in contents
and makeup, it is the same type of change we already undergo which might be more appealing
to some folks. You also presumably can’t easily hack or
alter it, in a computer sense, obviously you can hack it in a more traditional way but
it’s presumably much harder to smash a brain in some armored support vat then a classic
skull, or the body supporting it. I’m not sure why the brain in a jar or vat
is always shown as fragile-looking glass, but presumably that’s just the artist trying
to make its contents obvious. Your brain is already heavily armored compared
to most of your body and if you’re removing it from that body and skull, you are presumably
putting it in something a bit more solid than a glass mason jar. Of course we sometimes see them with the spine
or eyeballs still attached and a nice window certainly is helpful in that case, though
it is a reminder that our mind is not just our brain. There’s no non-arbitrary border to your
mind really, yes the skull is where the brain is kept but that’s not really the whole
of your actual mind from a thinking perspective – you’ve got that nervous system throughout
your body plus all those hormones produced by this or that gland. Those presumably can be replicated by having
small tanks of artificial hormones, or machines that produced them, injecting them into the
brain along with the normal blood and oxygen. Still, we probably shouldn’t assume someone
who wants to preserve their brain would automatically be fine with synthetics and if not require
a whole body, some folks might draw their lines at different places on what they preferred
so that you might have various glands in the nutrient bath too. Incidentally we’re just sticking with the
trope of a brain in a nutrient bath for simplicity’s sake, though cerebrospinal fluid is indeed
clear and colorless. We also have successfully removed pig brains
from their bodies and kept them alive, though not thinking or really experiencing thus far,
and we’ve had limited success with things like brain and head transplants and connecting
neurons to mind-machine interfaces, though needless to say ‘success’ thus far has
been modest and debatable, but techniques are improving. The concept though is not limited to simply
a brain being removed, but more the idea of removing all but the most vital components
needed to keep a person alive. Science and experimentation will help us figure
out what that actually is, and we’ll doubtless get that experience because folks will often
end up badly injured but with their brain intact, as it’s a small part of your body
that might get missed by damage and is heavily armored. While it’s neat to contemplate options like
brain scans and nanobot repairs, they are still quite a ways off and mostly not much
of interest to the public when it comes to funding and attracting professionals, whereas
we have plenty of folks who get gravely injured and might be able to live with these emerging
technologies, or have their life quality improved by them. So we’ll probably discover sooner than later
what we need to keep a person alive, and not necessarily in control of some sort of cloned
or android body, by which I don’t just mean an alternative like some giant mecha or war
robot or an inhuman form, like having your brain grafted into an animal or alien or a
spaceship as a body. We might keep a brain simply in a deep slumber
state or dreaming, as mentioned those initial big brain removals kept the brain alive but
with little to no activity so it might be - given that neurons live far longer than
most cells – that this approach of suspended or low activity thinking might become a more
practical alternative than freezing brains – which damages stuff. Or it could be done for experimentation or
as a punishment, keeping folks conscious in sensory deprivation or feeding them fake stimuli. This latter notion is the basis of the Brain
in a Vat thought problem, which people tend to assume is identical to the situation we’d
be dealing with in examples of digitally uploaded or simulated minds. One big difference between those is that with
a digital mind, uploaded or simulated, not only do you control all the sensory input
that mind experiences but you can pause your simulation or edit that mind or revert to
an older saved state or copy if you make an error that gives the fake reality away. You probably could technically do that with
an actual meat brain too, the line between organic and artificial or mechanical gets
hazy to the point of irrelevance with sufficient technology, but for the moment we’ll assume
all you control in a brain in a vat is the sensory inputs and of course the ability to
shut it off permanently. What is the actual Brain in a Vat thought
experiment though? It’s essentially a modernized version of
Plato’s Allegory of the Cave and Descartes’ Meditations on First Principles, such as ‘I
think therefore I am.’ Same basic principles involved as in simulation
based thought experiments but again differing in that the programmer is not limited to sensory
control of the person’s mind, they can edit that too, whereas the mad scientist with the
brain in a vat can’t. As we’ve discussed in episodes on that topic
before, like Simulation Hypothesis, Reality & Simulation, and Virtual Worlds, my own opinion
is that this tendency to try to delineate the real world from the fake one is often
not really a productive line of discussion, it just leads to solipsism, and hinges on
an arguably false dichotomy. Defining something as unreal because it’s
artificial is a pretty dubious notion. It might be a bit semantic to say that living
in a house whose architecture and décor you’ve selected – or somebody else did – isn’t
really much different than living in virtual environment you or someone else made, and
really seems to hinge more on the notion of not if the reality you are in is created but
rather if your existence in it is deceptive, and that’s more about truth and free will
than reality. As an example, we have theories that suggest
black holes might spawn new Universes when they form, essentially making those a reality
beneath our own, but also implying we might have been spawned by the Big Bang as a product
of a black hole forming in an older Universe. I’m not sure how that would make us less
real than the folks living in that Universe, or more real than one spawned by the formation
of a black hole in our Universe, or how any of that differs from a simulated Universe
beyond the implied artificial origin via a programmer or creator. That brain in a vat might exist as a separate
and real thing in the universe that’s running the simulations it experiences but so too
would a hard drive and presumably for the black hole example, that black hole in the
older Universe. Let’s say our technology allowed us to rescue
folks who were severely injured by placing their brains in vats and we lacked good android
technology so we stuck them in a simulated village and told them that was the case and
that they could call their friends and family in the outside world and that they had control
of that environment within the limits of their group decisions and the software’s limitations. That suddenly seems a lot less sinister and
philosophically interesting than the normal brain in a vat and I’m not really sure how
that place is any different than a remote village on Earth that’s cutoff from most
external contact beyond telephone and internet. You can’t know if the reality you’re in
is the one and only real place, and I’m not sure that it matters in and of itself,
as opposed to the question of if you’re in there voluntarily or aware of your circumstances
or if the folks controlling it are good or evil. Sometimes the better question isn’t if what
you experience is real or not, or how you’d know, but if it actually matters. When it comes to what life would be like as
a brain in a jar, much like a digitally uploaded mind, and depending on your technology, it
might be identical to normal life. Indeed with the right technology it might
be identical to being digitally uploaded, you can mess with a brain like you can a computer
file with enough tech, altering or reversing memory inputs in the meat of the brain. But we shouldn’t assume a programmer could
casually screw with a digital mind which is likely to be a very complex and sensitive
thing that might be severely damaged by minor edits, especially if it’s just an emulation
of a meat-brain. You could have a case where that brain is
just in an android body, which is arguably just a cyborg, it might have cybernetic enhancements
to the mind too, or have genetic tweaking to allow it to grow bigger. Classic sci-fi is full of giant-headed aliens
with highly evolved giant brains, and while newer stuff tends to focus more on the digital,
this augmented approach might turn out to be a more preferred route for transhumanism
or posthumanism, since it doesn’t involve copying your brain onto a hard drive or necessarily
adding mechanical or electronic components to the actual brain itself, just the sensory
and life support bits. And there’s a good chance this pathway will
be available to us a lot sooner than all the digital options, which I sometimes think only
seem more probable by embracing a handwave that mind uploading is really just about having
a ton of computational power and a very good brain scanner. So what are your options for life as a brain
in a jar? Well if you’re not tinkering with a basic
brain much you probably need an android body or a simulated one in virtual reality that’s
very human in function and inputs, but even assuming we can’t tinker much with that
actual brain to allow it to comfortably run inhuman forms, the brain itself is very adaptive
so folks might be able to adjust to running something more animal or vehicle-like instead
of human form. Personally I’ve very fond of the notion
of being in a giant human mecha, with gatling guns and missile racks, but the neat thing
is you could transplant the brain to other bodies, whether by moving the brain jar or
just do everything by telepresence with the brain in some nice armored bunker somewhere. The other aspect is that it might become a
preferred thing for civilizations. Bodies take a lot of calories to run, and
brains are big users of that, chewing up around 15% of most folks calories while making up
only around 2% of their body mass. So unlike digital minds, which might run thousands
or millions of times more efficiently than human brains, these would only let you get
maybe 6 times as many people for the same amount of food as normal folks like you and
I eat. Still it is six times as many, which is nothing
to sneeze at if you’re civilization wants to maximize its population. And there’s a couple caveats too. First, it’s only theoretical that we can
make hyper-efficient computers, we have made some supercomputers that match our best guess
for a human’s analogy of processing power but those things are power and heat gluttons,
sucking up many thousands of times more power than a human brain does. We probably can do way better in the future
than our meat-brains in terms of power for computation but shouldn’t just take that
for granted and assume we’ll eventually hit some super-high efficiency rate. And that only matters if folks feel a digital
mind is equal or better than going brain in a jar. Second, our brains use about 16 watts of power
to run, but that does not mean that needs to come from food grown on open land or in
hydroponic farms. Once you’ve abandoned the flesh in favor
of synthetic sensory input, worrying about things like taste and dietary fiber and texture
start being rather irrelevant, you do not have an actual tongue or stomach after all. None of the stuff our cells run on, be it
normal cells or neurons, really requires a plant grown under sunlight to make food we
eat directly or further up the food chain. There’s lots of things you need to feed
the brain but the big ones are glucose and oxygen and either could be introduced into
blood via synthetic production like photovoltaics, which is way more efficient than photosynthesis. If we just assumed we could only achieve a
10% efficiency of photovoltaic sunlight capture to glucose energy, you still wouldn’t need
a solar panel much bigger than the jar for the brain to keep it running and you could
have a planetwide brain farm running on solar containing over a quadrillion people. Nor would that necessarily need to be on Earth
either. Obviously we could use our usual notion of
rotating habitats but brains in a jar might not actually need gravity and if they did,
probably would do fine in a relative small centrifuge, since it probably would only need
low gravity and even if didn’t you can probably ignore most of the vertigo and nausea issues
associated to high-RPM rotation, what with the lack of natural ears and such. In such a case a Dyson Swarm created around
this purpose, somewhat akin to a Matrioshka Brain, allows far more people than a regular
Dyson Swarm. Those usually permit around a billion times
more people than could live on Earth and are very raw material intensive, requiring the
disassembly of an entire solar system or more – something we’ll look at more next week. Alternatively, a Brain-Jar-Dyson-Swarm, using
the 10% efficiency we suggested a moment ago could house 2.4 trillion-trillion people,
or brains, and probably would only need about an Earth’s worth of raw materials to construct. I’m not sure many folks would find that
appealing but it might be entirely fine with folks, if it was accompanied with virtual
worlds and telepresence and allowed thousands of times more folks to live and enjoy such
places. Indeed they might find it a better and safer
and longer-lived alternative. I’m not sure it would be my cup of tea,
but as I said earlier, while continuing as I am is my current preference, the notion
of stalking around the landscape as brain inside a giant robot has a lot of appeal. We were talking a lot today about brains versus
artificial intelligence or mind uploading, and if you’d like to learn more about concepts
like neural networks or computational biology, there are some fantastic courses on those
topics over at Brilliant. To truly understand the human mind, or to
devise artificial intelligence, we’ll need to develop a deep understanding of Neural
Nets and they are a growing field in academics and business and Brilliant’s focus on fun
and interactive methods makes them a great choice for learning about this topic or many
others, whether you’re a student, a parent trying to enhance your kid’s education,
a professional brushing up on cutting-edge topics, or someone who just wants to use this
time to understand the world better, you should check out Brilliant. Try adding some learning structure to your
day by setting a goal to improve yourself, and then work at that goal just a little bit
every day. Brilliant makes that possible with interactive
explorations and a mobile app that you can take with you wherever you are. If you are naturally curious, want to build
your problem-solving skills, or need to develop confidence in your analytical abilities, then
get Brilliant Premium to learn something new. Brilliant’s thought-provoking math, science,
and computer science content helps guide you to mastery by taking complex concepts and
breaking them up into bite-sized understandable chunks. You'll start by having fun with their interactive
explorations, over time you'll be amazed at what you can accomplish. If you’d like to learn more science, math,
and computer science, and want to do it at your own pace and from the comfort of your
own home, go to brilliant.org/IsaacArthur and try it out for free. So this weekend we have a bonus episode coming
out, the first installment of a new series, Becoming an Interplanetary Species, where
we will look at the National Space Society’s Roadmap to that Stars, and see what the first
steps are to us becoming an Interplanetary Species. Last week we looked at some truly enormous
space habitats we might build in our future as an alternative to settling new planets
and next we will take a look at how we can go about acquiring the vast amounts of raw
materials we’ll need to construct millions of those continent sized habitats, and ask
if we should dismantle the solar system itself to provide them. After that we’ll be back to the Fermi Paradox
series to consider disappearing stars and cosmic voids, to consider if such things are
natural or might be signs of older alien civilizations dismantling their own solar systems or even
entire galaxies. If you want alerts when those and other episodes
come out, make sure to subscribe to the channel, and if you’d like to help support future
episodes, you can donate to us on Patreon, or our website, IsaacArthur.net, which are
linked in the episode description below, along with all of our various social media forums
where you can get updates and chat with others about the concepts in the episodes and many
other futuristic ideas. Until next time, thanks for watching, and have a great week!
