Sam Harris speaks with Christof Koch about the nature of consciousness. They discuss Christof’s development as a neuroscientist, his collaboration with Francis Crick, change blindness and binocular rivalry, sleep and anesthesia, the limits of physicalism, non-locality, brains as classical systems, conscious AI, idealism and panpsychism, Integrated Information Theory (IIT), what it means to say something “exists,” the illusion of the self, brain bridging, Christof’s experience with psychedelics, and other topics. If the Making Sense podcast logo in your player is BLACK, you can SUBSCRIBE to gain access to all full-length episodes at samharris.org/subscribe. Learning how to train your mind is the single greatest investment you can make in life. That’s why Sam Harris created the Waking Up app. From rational mindfulness practice to lessons on some of life’s most important topics, join Sam as he demystifies the practice of meditation and explores the theory behind it.
Just a note to say that if you're hearing this, you're not currently on our subscriber feed, and will only be hearing the first part of this conversation.
In order to access full episodes of the Making Sense Podcast, you'll need to subscribe at SamHarris.org.
There you'll also find our scholarship program, where we offer free accounts to anyone who can't afford one.
We don't run ads on the podcast, and therefore it's made possible entirely through the support of our subscribers.
So if you enjoy what we're doing here, please consider becoming one.
Today I'm speaking with Christoph Koch.
Christoph is a neuroscientist at the Allen Institute and at the Tiny Blue Dot Foundation.
He's the former president of the Allen Institute for Brain Science and a former professor at Caltech.
He's the author of five books, most recently, Then I Am Myself the World, What Consciousness Is and How to Expand It.
Which is the topic of today's conversation.
Christophe and I speak about his development as a neuroscientist, his collaboration with Francis Crick, their studies of the visual system, change blindness, and binocular rivalry.
The significance of sleep and anesthesia for consciousness studies, the limits of physicalism, non-locality and other quantum mechanical phenomena, brains as classical systems, the possibility of conscious AI, idealism and panpsychism, integrated information theory, Also known as IIT.
What it means to say that something exists.
The illusion of the self.
The possibility of brain bridging, that is connecting two human brains in a shared experience.
Christoph's recent experience with psychedelics and other topics.
Anyway, this is the deep end of the pool with respect to the scientific understanding of consciousness, and I certainly enjoyed it.
And I bring you Christoph Koch.
I am here with Christoph Koch.
Christophe, thanks for joining me.
Thanks, Sam, for having me.
So, you've written a wonderful book.
You've written several books, but your most recent is Then I Am Myself the World, What Consciousness Is and How to Expand It.
And I just really want to follow the line you took in your book here, which traces your The evolution of your thought as a scientist focusing on the nature of consciousness.
And you've had a very productive career as a neuroscientist and you've had a very fruitful collaboration with Francis Crick, who we'll talk about.
But then you, you know, kind of late in your career, you took the first-person side of things with both hands and have had some experiences with psychedelics that have Put pressure on your ontology, one might say.
So, I would just love to talk about all of this.
Perhaps you can start somewhere near the beginning.
How is it that you came to focus on consciousness?
You started out more as a physicist.
That was the side of science you came in at, I believe, as an undergraduate.
What led you to the study of consciousness?
I did get a minor in philosophy in Tübingen, which is a 550-year-old university in Germany.
And I did grow up reading Schopenhauer and Immanuel Kant.
It was that sort of household.
So I've always been interested in consciousness.
That voice inside the head, right?
How is that voice compatible with everything else we know?
And I grew up reading physicists, including Ernst Mach and people like Schrodinger, and they all wondered, they also had similar questions.
And in fact, both Schrodinger and Ernst Mach were very explicit, before I can be a physicist, I am a conscious being that sees that things, you know, going back all the way to your renegade cloud.
And so before I can even read an oscilloscope or read an instrument as a physicist, I depend on my conscious sensation.
And so consciousness has to be at the center of of our explanation of everything in the world.
And I was puzzled when I first came from physics into neuroscience.
So my PhD was in the 20th century in what we now would call computational neuroscience.
I was puzzled when I really went deep and became a full time professional neuroscientist that consciousness at the time was simply not discussed.
It didn't figure in the index.
You know, if you go, if you got the standard textbook, you went index on a consciousness, nothing.
That was simply ignored.
It was all about behavior and neurons, which is fine, which is what I dedicated most of my life to.
But ultimately, we also have to explain the puzzle at the heart of our existence, the fact that I'm not just a behaving thing.
I actually see, I hear, I dread, I fear, I dream, I desire, I want.
So how do all of these things get into the world?
And then I met Francis Crick.
So I did my PhD in Germany, then went to do a postdoc at the Artificial Intelligence Lab at MIT.
And previously, I'd encountered Francis Crick, you know, who at the time had shifted from molecular biology where, you know, with Jim Watson, he decoded the molecular code for life and deciphered the double helical shape of the DNA molecule.
But then he shifted because he was also interested in consciousness.
And the two of us got together and then wrote roughly 20 papers over the next 14 years where we really initiated sort of a modern empirical program to discover the footprints of consciousness in the brain.
And the argument was very simple.
Never mind about all the philosophy.
We aren't going to converge anytime soon.
You know, are we a physicalist, an idealist, a dualist, a panpsychist, whatever.
Let's just focus.
We all agree today it's not the heart, as most people thought, it's your brain that's sort of the organ of consciousness.
So, which particular bits and pieces of the brain?
Is it the spinal cord?
Is it the cerebellum?
Is it the retina?
Is it the thalamus?
Is it the cortex?
Do they, does this bit of the brain that sort of analyzes the substrate of consciousness, does it have to oscillate?
Does it have to buzz at a particular frequency?
Which genes are involved?
Which cell types are involved?
Is there sort of a conscious mode and an unconscious mode?
These are all questions that we felt Will have an answer.
No matter what your philosophical predilection is, there will be an answer of the sort.
These neurons in this state in that part of the brain at this time express the fact that I see or that I hear.
And it's causal.
In other words, if I can activate those neurons, if I can put these neurons into this particular buzzing state, for instance, then you should have a conscious experience.
Conversely, if I can somehow interfere with this neural mechanism by a drug, by a neurosurgeon's electrode, by some external magnetic device, then you shouldn't have the experience.
And so this is sort of the modern program to study the footprints of consciousness now called the Neuronal Correlates of Consciousness or the NCC in patients, in neurotypical volunteers, in animals like mice and monkeys and rats and other animals.
But when I read you now, I recognize that you are someone who always took the so-called hard problem of consciousness seriously, which is to say you thought it was a non-trivial mystery that it should be like anything at all to be associated with a certain pattern of neural firing.
Whereas, in my memory of Francis Crick's approach to this, I mean, he wrote a book, The Astonishing Hypothesis, way back when.
I remember him being more of an arch, you know, materialist, you know, of the sort that, you know, one met quite directly in my friend Dan Dennett or in the Churchlands, right?
Somebody who just simply tried to ram past the hard problem with the Declarative statement that, you know, the mind is simply what the brain is doing, right?
You're just a pack of neurons.
And there is no, you know, it seems like there's something left out of that.
That's just a symptom of what a bad view we have of the brain's role in producing subjectivity.
But, so how did, from what you just said, it sounds like you sort of tabled your philosophical differences of intuition there and just decided to go look for the neural correlates of consciousness.
It must be somewhere in the brain.
We agree about that, whether we agree about anything else in philosophy.
But was that an impediment to your collaboration with Crick at all, or do I have Crick wrong?
No, you don't have it wrong.
That was the starting point.
That was the explicit starting point that once we understand neurons and their vast complexity, you know, of an untamed complexity that's incomprehensible for us, you know, humans, we deal with simple networks that are, you know, that we can understand, but we've never been faced with the vast complexity that we find in evolved systems like
Like biology, whether that's the brain of a simple creature like a worm that has three or two neurons, or the brain of a human that has on the order of 100 billion neurons.
But he was also, Francis was very clear, he said, this hypothesis may not be true.
There might be other ways we have to think about it.
And he was sympathetic.
So for instance, very early on, we encounter Jerry Edelman, another Nobel laureate who also moved from Manhattan From the Rockefeller Institute to the Scripps Institute in La Jolla, and he worked with a person at the time, Giulio Tononi, now also a very well-known consciousness researcher, and he explored the possibility that maybe it has to do with a complexity.
They wrote, for example, an early influential paper called Consciousness and Complexity, arguing that complexity had to be involved, which is a little bit more than just saying it's just a bunch of neurons.
You know, because the most widespread belief among neuroscientists is, well, it's an emergent property, just like wetness emerges from water.
You don't get, if you have two H2O molecules, they're not wet, but if you've got, you know, 10 to the 23, like a liter of H2O molecules, then it gets wet.
And similar, if you have a few neurons, they're not conscious, but you've got 100 billion of them, then somehow they're conscious.
But then we also, Francis and I, realize that's inadequate because you have some structures like the cerebellum, okay?
So you have this little brain tucked underneath your big brain, right, at the back of your head.
It contains, in fact, 80 percent, four to five neurons in your head and the cerebellum.
You can lose these neurons, let's say due to a stroke or due to a tumor.
You will be impaired.
You can't do fast speed typing on your phone anymore.
You can't play violin or piano anymore.
You have a few other issues like that, but basically you stagger about.
You look like you're always drunk.
But basically, all these people, these patients who have lost part or whole of their cerebellum, they see, they hear, they dread, they fear, they imagine, their consciousness is essentially to first order unchanged.
And so that tells you that it can't just be the number of neurons.
It has to do with at least with the way they're organized.
Same thing with the spinal cord, right?
You can be quadriplegic.
You've just lost all your spinal cord, 200 million neurons, so you can't move.
But again, your consciousness hasn't really changed that dramatically.
So it can't just be the number of neurons, it has to be the way they organize.
And Francis, the hope was Similar to what he had accomplished in molecular biology, that if we look at the right neural mechanism in the right way, then suddenly it'll become apparent, just like it was apparent when you look at the double helical structure of DNA, that that's a natural way to copy genetic information.
But we, of course, we didn't find such a simple explanation, and he always was open to the possibility that other ways of conceiving of it may be necessary to finally understand it.
Well, I should say that anyone who has seen me try to type on my phone will wonder whether I have lost my cerebellum in some terrific accident.
Well, you know, a few people are born without a cerebellum altogether.
You know, the so-called agenesis of the cerebellum.
Well, I can confirm I do have one, having done my fair share of MRI experiments.
Oh, good.
There you go.
Actually, I don't know if you're aware of it, you and I once met, when I was doing my PhD at the Brain Mapping Center at UCLA, you came and did a seminar.
A long time ago.
Yeah, at Mark Cohen's lab.
Yes, a long time ago with Locke, yes.
Yeah, I don't think our paths have crossed since, but... No, no.
Okay, so now how has your... now your evolution as a neuroscientist in trying to understand consciousness as an emergent property of brain activity, you know, I remember the work you did with You and Crick were focusing on the visual system, if memory serves, and you were looking at things like bistable, percepts, and change blindness.
Binocular rivalry.
Right, right.
Binocular rivalry.
That's a fascinating phenomenon.
Maybe you could describe that and why that seemed like such a promising avenue, because to experience it is striking.
Both binocular rivalry and change blindness are quite wonderful to experience as a subject in an experiment.
Yeah, so the general philosophy here in these experiments is to create conditions where on the one hand, you're looking at the stimulus and see it.
And then when the same stimulus is present, you may not see it or you may see it differently.
In other words, in the two conditions, you're always looking at the same thing, but sometimes you see it one way and sometimes you see it the other way.
So in change blindness, you know, you have an image like there's a famous image of soldiers You and soldiers boarding a plane, and then you see an image that's slightly changed, with a blank in between, where, for example, the photographer has removed, using Photoshop, has removed the engine of the airplane.
And you go back and forth between the original image, a blank image, and the changed image.
original image, blank changed image.
And you may be staring at this for 10 or 20 or 30 seconds and your eye move about, but you don't see the actual change until, you know, suddenly you get, ah, of course, it's the engine that's changed.
And then, of course, you cannot not see it.
So there you have a situation where for 10 or 20 seconds, you're looking at the same thing, but you're simply not seeing what's in front of your eyes.
So you can track down where are the mechanisms in your brain that respond only to the retinal input, compared to where are the neurons that respond when you actually see it.
A separate case is this binocular rivalry.
So there you have two different images, let's say phase one and phase two.
Let's say you're projecting phase one into one eye.
To make it concrete, we can think about, for instance, you have an image.
You can do this.
We did it 20 years ago with the two presidents at the time.
You can imagine on your left image, you have Biden.
On your right eye, you project the image of Donald Trump.
Now, if you don't do things correctly, you can see sort of a juxtaposition of the both, of the two.
But if you do things correctly, for a while you will only see President Biden.
And then one or two seconds, or three seconds, and then Biden will fade and Trump will come in.
And then Trump will stay for several seconds, and then Biden will come in.
It's sort of this never-ending dance.
So just to clarify for listeners who may not be able to visualize this, you have one image projected to one eye and a different image to the other eye, and your conscious experience tends to, if it's set up correctly, is you're simply seeing one of the images, and then it just randomly shifts to the other image, and it's something over which you... You go back and forth, yes.
You can't control the shift, and what's interesting about this experimentally is that The inputs to your nervous system are completely stable.
I mean, all you have are these two inputs to, you know, one to each eye, and nothing is changing.
And yet, at the level of your conscious, you know, phenomenology, there is this very bizarre shifting back and forth between seeing one image and then seeing the other, and seeing one image and seeing the other.
And this is a wholesale change in the contents of visual consciousness, and yet there's zero change happening at the level of the bottom of your nervous system from the, you know, the retina onward.
That's correct.
That's correct.
So something is changing there, sort of this bistable.
It's like the nervous system wants to see what's in the left eye, and then it wants to sort of check what's on the right eye, and it goes back and forth, and this can go on for, you know, many minutes.
So now you can, in principle, You can do what you used to do.
You can put people in a magnet, or you can put people in a magnet like we've done at the time at UCLA with Isaac Fried on the wall.
These are neurosurgical patients that for other reasons, because the doctor has to know where the elliptic seizure originates, you put an electrode into their brain.
You can see the same thing, or you can do it in a In a monkey, where you can track throughout the monkey's visual system, where are the first stages of the neurons that actually respond to what you see?
Not just what the left eye quote sees or what the input received by the right eye, but what the animal sees or what the person sees.
That typically tends to be high up in the In the visual hierarchy.
And you can ask, where are those neurons?
Are they of a particular type?
Are they of a particular layer?
And then, of course, what happens if I can begin to manipulate those neurons?
Which at the time wasn't really possible, but today you can do using fancy optogenetics.
So this was one of a number of ideas that we exploited to track or that people exploit to track the neural footprint of consciousness.
In this case, visual consciousness throughout the brain.
And people continue to do that in all sorts of different ways.
Is there a problem here conceptually?
Because you're obviously conscious all the while.
There's a distinction we could make between consciousness itself, I mean the fact that it's like something to be you, to use Nagel's definition, of which I've always been a fan, And any specific contents of consciousness, right?
So, there you're really interrogating specific contents, but one could argue that you could completely lose a perception of the visual field as well and yet be just as conscious, right?
I mean, now you're no longer conscious of vision, but you're conscious in other sensory channels and of just awareness itself, one might argue.
It's a different way to study.
So the other way to study, as you point out, is the distinction between consciousness, too cool, and not being conscious, i.e., like in sleep, or in anesthesia.
So you can take a subject, normal subject, hears, sees, etc., then you anesthetize them, and you can see what changes.
And then as they wake up, it's called LOC, loss of consciousness, and then you can see what happens when they come out of anesthesia.
And you're looking at different things, and they're not similar, and there are different concerns you have to have.
So, for instance, when you go to sleep or when you're anesthetized, all sorts of other things change as well.
For instance, you don't have memory anymore.
You can't move anymore, right?
Because you become paralyzed in deep sleep or in anesthesia.
You also become paralyzed But it's much more dramatic compared to looking at something.
And as you point out, I'm still conscious.
It's just the content of consciousness changes.
So they're just different experimental techniques to track different things.
In one case, you're specifically tracking the content of my visual consciousness.
In the other case, you're tracking the entire physical substrate of being consciousness in the negative sense.
It feels like something to be me.
Yeah.
The problem you always run into there, which you just alluded to, is that it's hard to differentiate and perhaps, in principle, it could be impossible to differentiate a true cessation of consciousness from merely a loss of memory, right?
So, you know, I've... I disagree.
Yeah, so let's talk about that.
I mean, I think it's harder to make this case for general anesthesia, given what we think about just the underlying state of the brain.
But for deep sleep, I've often felt that we're too quick to allege that it interrupts conscious experience, because many people I count myself among them, at least believe, imagine, that they've had an experience of dreamless sleep.
Now whether that's, you know, stage four delta sleep or not, I don't know, but there's a very common experience of being asleep and dreaming and there's something that it's like to be doing that.
But there's a more esoteric experience that people have had of being asleep and experiencing just a... I mean, it's very much like a meditative, you know, samadhi-like experience.
I mean, it's just, you know, a vast experience of contentless consciousness while asleep, right?
And that's certainly reported among yogis who claim to have been able to train that.
So I just think, you know, that's the fact that most of us Most of the time, don't remember a thing about what it was like to be deeply asleep could be just analogous to most of us not remembering our dreams.
I presume I dream every night, but at this point in my life, I rarely remember anything from my period of sleep.
So, there is one technique, at least, that tries to get... I can think of two techniques to try to get at that.
One is to put you in a magnet, or probably more plausibly, to put EG electrodes on you as you sleep.
And people have done this.
I'm thinking of one study by Ciclari et al., from the Tononi lab, in fact, where they did exactly that.
They had high-density EG on these irregular people, but they're trained observers.
So, they've done this for a while now.
And you randomly wake them up.
So randomly, I look at your EG and I wake you up now and I ask you, you know, tell me, was anything in your mind just the moment before you woke up?
And I look at the 10 seconds before I wake you up.
And then I can correlate.
And of course, what they found, which is, if you can go back to the original report back in 53, defining REM and non-REM sleep, it is true.
30% of the time, even in deep sleep, people do report dreamlike experiences.
Okay.
So even in deep sleep, you can have dreamlike experiences.
However, they're typically not this sort of elaborate narrative structure.
You know, they're more sort of episode, oh, I just saw a face, or I just had this feeling of dread, something bad was going to happen.
And conversely, in REM sleep, only about 80% of the time do people actually report you know, have dream reports.
So that's one way to get around this fact that most people forget most of the dream.
You wake them up spontaneously during the night and you ask them, do you remember anything over the last 10 seconds?
Now, of course, it's also possible that if they have lost short-term memory as well, then of course, yes, then of course you wouldn't know it all.
But most people can't report that they had experience, dream experience, or that they didn't have dream experiences.
And here, by the way, so you can, again, People have used this technique to track the footprints.
It is true when you have dream experiences, it's typically a particular part of the back of the brain, the posterior heart zone, that are activated in the preceding 10 or 15 seconds by EG criteria.
We're talking the posterior hot zone.
You're not talking about visual cortex.
You're talking about precuneus and… Yeah, exactly.
Precuneus, posterior cingulate, sort of parietal areas, and some higher order temporal areas.
So, behind the central fissure that goes across the brain, in the back.
So, let's step back for a moment and review some of the underlying philosophical controversy, because there are many ways in which people think about what is here, and I want us to get to integrated information theory, IIT, which I believe you still are a supporter of and contributor to.
I don't know if it fully captures your current view of things.
But we'll get there.
But you express in the book a similar astonishment and frustration that I've always felt with the eliminative materialists of the Dan Dennett and especially Paul and Patricia Churchland sort who just simply say that consciousness itself is an illusion
It seems to me, it's always seemed to me to be absolutely obvious that consciousness is the one thing about which you cannot say that, because any illusion is as much a demonstration of consciousness as any veridical perception of anything, right?
I mean, it's the one thing.
The fact that something seems to be happening is the one thing we can be sure of, even if we're totally confused about everything.
René Descartes.
I couldn't agree more.
It's one thing I cannot doubt because in doubting I affirm it.
And if I call it an illusion, well then everything is an illusion and the word becomes meaningless, yes.
So I think this is the biggest failing by far of physicalism, right?
So physicalism, you know, it's a metaphysical idea that the only thing that exists is physical.
Let's come, you know, I think we have to discuss what is meant by that.
But sort of most people have an intuition, you know, it's matter and energy, right?
Good old materialism.
And then the challenge is, well, if you believe that, fine, but then how do feelings emerge?
And, you know, we have been not able, you know, philosophy has been utterly unable, or science has been utterly unable to explain how any sort of feelings, it feels like something to be me, emerges out of atoms and the void.
That's the biggest challenge that physicalism has utterly failed to meet, number one.
Yeah, which is to say, so again, I've spoken about the hard problem of consciousness a lot on my podcast and over at Waking Up in the app, and I've interviewed Chalmers and
Anil Seth, and we've covered this ground before, but the thing to recognize is that the fact that there is this explanatory gap, the fact that our intuitions that seem to anchor every other type of scientific explanation, you know, the wetness of water, the brittleness of, you know, any higher level material,
All of those physicalist reductions run through, and yet, with the hard problem, with the fact that the lights are on subjectively, something seems to be left out.
The fact that we have this intuitive impasse doesn't suggest, or much less prove, that it isn't simply so, right?
That it isn't just... We just may be in a bad position to think about How consciousness emerges, and it may always seem like a miracle, even if we had the answer in hand, and it just was, whatever, 40 Hz oscillations in thalamocortical loops, which I think was once a pet thesis of yours.
Yes.
If that's just the answer, if God told us that is how it happens, well, it may always just seem like a brute fact that doesn't actually explain anything.
You're entirely right.
My dog doesn't understand general relativity or the stock market or election.
My dog is a perfectly intelligent member of a species that survived for millions of years, but certain things are cognitive beyond it.
And maybe this is also beyond us.
But now the additional evidence for physicalism being inadequate, namely at the bottom, at the rock bottom of physicalism, is how do we define the physical?
And if you listen to anything in quantum mechanics over the last 30 years, we all know it's deeply troubling and it's very difficult to define what is the physical.
And the fact that physically includes such bizarre things as two particles that are entangled, that are at opposite ends of the universe, if one, you know, if you observe one and determine its state, instantaneous, you know, instantaneous the state of the other one is determined.
You're talking about non-locality.
Non-locality.
So what sort of, I mean, what sort of physicalism is that if things are entangled across the universe, right?
That's certainly not my grandfather's.
That's certainly not Democritus, you know, atoms in the void.
And then, you know, now it turns out that the entire school of physics, you know, that does what's called contextuality or called first-person physics, right, where it is, where they accept as a fact, as an observational, as an empirical fact that what exists really depends on what you measure.
And if you have different measurement protocols, different things sort of, you measure different things that weren't there before.
So the mere act of observing, you know, it's a participatory universe, the mere act of observing creates reality.
Well, how does that sit with standard sort of physicalism, right?
So there's a All right, let me put on my arch-materialist hat, which I rarely wear, but what about the claim that however strange quantum mechanics may be, and I guess there's two things to acknowledge here.
One is that there are very strange versions of it that seem to move consciousness and the observer out of any part of the significant picture and give us something... Super-determinism!
Or, you know, many-worlds theory, right?
Where, like, you know, the universe is split in, you know, at every possible choice point, of which I think... But how do we know this exactly?
So, in fact, I've just written a paper with Hartmut Neven, who runs Quantum Google AI Lab, It may be precisely consciousness that's responsible for, at the moment you create a superposition, there are two or a multitude of different multiverses that split off.
So, how do we know that conscience isn't, in fact, responsible for that splitting off?
I'm going to have you fight with David Deutsch about that, but Well, leaving that aside, what about the claim that at the scale of brains and even at the scale of neurons, kind of the wet and hot mess of goo that is in our heads, quantum mechanics, you know, is a perfectly classical view of physics is good enough because everything, you know, everything decoheres and there's just, you know, it's at the scale of brains, none of that highfalutin physics need apply.
Yeah, in short, brains are warm and wet.
So this is the argument I've always made in all my previous books.
The brain is warm and wet.
So if you look at Google's quantum computers, they operate at 23 milli Kelvin.
So that's about a thousand times colder than the temperature of outer space, which is the temperature of background radiation, a couple of degrees of Kelvin.
And of course, a hundred thousand times colder than right now, what you and I experience here on the West Coast.
So, in my belief, it's always been, yeah, it's wet.
The brain, as you said, is warm and wet, and so it's irrelevant.
But now we learned over the last 10 years, well, they can find entanglement in all sorts of soft metal systems, in some gas.
There was a paper recently out of Barcelona, I think, that measured it in gas at sort of minus 50 Celsius.
So people are now, as they're looking, as we better understand in this, what some people call the second revolution in quantum mechanics, it may not be true that it's only in these very extremely cold systems that entanglement and superposition may occur, even in what physicists think of soft matter, i.e., you know, brains or bodies.
This is an empirical question.
In fact, together with some biologists and with this previous mentioned Hartwood-Navin, we're now doing some experiments in organoids and in flies to exactly test it.
There's this interesting phenomena that xenon, so xenon is a rare gas, And it acts as an anesthetic.
This is well known.
In fact, it's a pretty good anesthetic, except it's expensive.
But there's this one paper many years ago in mice that claimed there's this differential isotope effect of anesthetic potency.
In other words, different isotopes of xenon.
So, 129, 131, 132, they all have the same, so chemically they're all the same.
They all have eight outer electrons, so none of them interact.
They're called Nobel gases because they're so noble they don't interact with anyone.
So, only what differs is the nucleus inside.
They have an additional one neutron, or two neutrons, or three neutrons, and so their nuclear spin differs.
Some of them have nuclear spin zero, some of them have nuclear spin one-half.
And it turns out those that have nucleus spin one-half have a different anesthetic potency in mice—it hasn't been done in people yet—in mice than the ones with spin zero.
So this has to be replicated, right?
Two out of three experiments in biology can't be replicated.
So we have to repeat this, which is what we're doing in flies and in terrible organoids and in primary cultures.
with Ken Cosek, who's a professor at UCSB, if this is replicated, then it would seem to indicate that something as subtle as nuclear spin actually makes a difference at room temperature in these organisms.
So the great thing about science-- - That would be very weird.
It would be weird.
And that's a great thing about science.
If you ask nature the right question and you listen carefully, you can get an answer, even if the answer may be extremely unexpected.
So that would be cool.
We'll see.
Yeah, yeah.
Well, keep me apprised of that because that would shake some assumptions I have working on the background here.
Me too.
So let's talk about, again, I want us to get to IIT, but how would you characterize the most common ontology of mind in neuroscience and the sciences of mind generally now.
I think in your book, you describe it as computational functionalism.
It's certainly the view that is causing people to imagine that developments in AI have some future implication for how we understand Minds in their totality and that we may build conscious machines.
We may be able to upload ourselves into the matrix.
All of this seems to suppose that mind, we already know this is true of intelligence.
I mean, it's clearly intelligence on some level as a matter of information processing, but the underlying thesis seems to be that consciousness itself is an emergent property of information processing and therefore really by definition is substrate independent Is that an accurate view of what you think most people think in the sciences of mind at the moment?
Yeah, based on some version of functionalism.
Everything has a function.
Clearly, that's true for intelligence.
Consciousness also has a function, you know, enable planning or, you know, long-term planning, short-term planning, summary.
Of the current situation that's ongoing in my body, whatever it may be, it has one or more functions.
And once you replicate those functions, in particular on a Turing machine, this is called Turing functionality or computational functionalism, then that machine will have all of the properties, including consciousness.
So that's a deep belief, particularly in the tech industry and among computational neuroscientists.
And now, of course, with LLMs, It's risen to the level of the general public that people assume, yeah, these things sooner or later will be sent in, will be conscious.
Right.
Here is what I predict is going to happen, and it worries me on some level.
I think we will probably build humanoid robots that fully pass the Turing test and better because they'll be superhuman in virtually every respect.
And once we get out of the uncanny valley with them, they'll certainly seem conscious to us because we'll build them to seem conscious to us.
And they'll talk about their experience and their emotions, etc.
They'll certainly be very attentive to our emotions and better judges of that than probably any person we've met.
And we will just effortlessly slide into this sense of being in relationship to these entities.
And we will still not understand the neural basis of consciousness or the computational basis of consciousness or any other basis of consciousness.
And we'll be in the presence of these artifacts where...
Which seem conscious to us and we'll simply just lose sight of whether it's an interesting problem to wonder whether or not they are conscious.
We will helplessly perceive them to be conscious because they will seem that way.
Can you imagine that we're going to stumble into that sort of Westworldian future where it's just You're certainly not going to be able to mistreat these robots because you'll feel like a psychopath, because only a psychopath could want to mistreat something that is so perfectly seeming to be a locus of experience.
It just seems like the problem may evaporate for most people.
I mean, obviously, some people will still hold on to it and wonder, Whether these machines are conscious, because obviously it'll be very important ethically to understand whether we have built machines that can suffer.
You know, if the lights are on over there and, you know, we're basically murdering our robots every time we turn them off, that will be an interesting problem to have created for ourselves.
But I just think if the robots, if we're out of the uncanny valley before we actually understand how consciousness is integrated with the physics of things, We just might lose sight of the problem.
For a short time, for a short period, this may happen, yes.
But I don't think it's stable.
I don't think it's stable.
It's a stable situation, inherently.
Because these intelligences will evolve at a timescale that's simply not matched the timescale that human society or human as individuals has evolved.
And so they rapidly will surpass us.
And of course, if we believe they are conscious, and they, as you point out, they have all the moral attendant responsibilities and rights, It'll further dehumanize us and will further dehumanize nature and will become less and less relevant because, you know, we're building these other guys.
They are our successor.
It's obviously the next step beyond humanity.
They're smarter than us.
They're more robust than us.
Ultimately, they're more aggressive than us because we'll build them for warfare.
You can see that already, you know, beginning to happen in Ukraine and Russia.
And where does this leave Homo sapiens sapiens?
Yeah, but for most people this problem will go away because of course they're conscious.
How can you not?
You can talk to them.
People will have intimate relationships already now, right?
You can talk to these companies who make these mobs.
If you'd like to continue listening to this conversation, you'll need to subscribe at SamHarris.org.
Once you do, you'll get access to all full-length episodes of the Making Sense Podcast.
The podcast is available to everyone through our scholarship program, so if you can't afford a subscription, please request a free account on the website.
