A 25 year-old bet on human consciousness gets settled
25 years ago, a neuroscientist and a philosopher got talking about consciousness and how to detect it in the brain.
“We made a bet about the nature of the neural correlative of consciousness,” David Chalmers, a professor of philosophy, says.
And who won?
“Fast forward 25 years. We are at this big event in New York,” Christof Koch says. “I conceded publicly on stage that he had won the bet, and so I gave him a case of six fine wines.”
Today, On Point: The continuing hunt for the home of the human soul.
Christof Koch, meritorious investigator at the Allen Institute for Brain Science. Chief Scientist at Tiny Blue Dot Foundation.
David Chalmers, professor of philosophy and neural science and co-director of New York University’s Center for Mind, Brain and Consciousness. President of the American Philosophical Association Eastern Division. Author of “Reality+: Virtual Worlds and the Problems of Philosophy” and “The Character of Consciousness.”
MEGHNA CHAKRABARTI: We exist. We eat, drink, breathe. We watch and listen. We plan things through. Such as, “When she steps out of the elevator, then I’ll step in.” But a robot could do a lot of that. What’s different is that we watch ourselves as we do these things, we think about ourselves and everything around us.
We feel waves of different emotions as we move through our lives, and we call that consciousness. So central to our existence, and yet nobody really knows what consciousness actually is. Christof Koch is a neuroscientist who has devoted his career to looking for traces or footprints of consciousness in the folds of the brain.
David Chalmers is a philosopher. He, too, has devoted his life to figuring out what consciousness is. The two of them have a great story to tell about their quest. It involves expensive wine, some good laughs and a lot of big questions.
And Christof Koch joins me now from Seattle. He is president and chief scientist at the Allen Institute for Brain Science, as well as chief scientist at the Tiny Blue Dot Foundation.
Christof, welcome to On Point.
CHRISTOF KOCH: Thank you Meghna. Just one correction. I am now, I stepped down for my executive duties. I’m now just a regular scientist at the Allen Institute.
CHAKRABARTI: Ah, okay, wonderful. So I will honor your humility there and be sure not to refer to you as president. But also joining us now is David Chalmers.
He’s a professor of philosophy and neuroscience and co-director of New York University’s Center for Mind, Brain and Consciousness. And author of “Reality+: Virtual Worlds and the Problems of Philosophy.” David, welcome to you.
DAVID CHALMERS: Thanks. It’s great to be here.
CHAKRABARTI: I wonder if you could both first take us back to what, 1988. And was it a late-night conversation in a bar that gave rise to a very peculiar bet, David?
What was happening that night?
CHALMERS: Yeah, this, I guess it was 1998, pretty much 25 years ago. We were both attending a conference, the second conference of the Association for the Scientific Study of Consciousness, a group which had been gotten together consisting of neuroscientists, psychologists, philosophers, trying to understand consciousness.
And the focus of this meeting was the neural correlates of consciousness. That is the areas of the brain which are active when you have a conscious subjective experience, that feels like something from the inside. And Christof, maybe eight years previously around 1990, along with Francis Crick, had articulated the search for neural correlates of consciousness as the centerpiece maybe of a growing scientific attack on consciousness.
To try and understand it. Find out which bits of the brain are active when you feel pain, when you see something red, when you think of a thought about your mother. And at this meeting, he was very optimistic about it. We had just been discussing the neural correlates of consciousness. He says, “I think we’re going to find the neural correlates of consciousness within 25 years.”
I said not so fast. I think we may well discover neural correlative consciousness eventually, but I think it’s a tough problem. It’s going to take longer than that. So we ended up making a bet.
CHAKRABARTI: Christof, can I just ask you, it was fascinating to me to see that you had worked with Crick, right?
So because the discovery of the double helix of DNA really unlocked our understanding of so much of about life. So perhaps, is that one had given you the confidence that science would eventually similarly unlock the secrets of consciousness?
KOCH: Fundamentally, I thought as a scientist, the central aspect of our lives, as you pointed out in your introduction, are our conscious experience of the world.
The only way I know, in fact, that I exist, is because I have feelings. I have experiences, right? This is the central assertion of René Descartes’ “Cogito, ergo sum.” And so I felt strongly together with Francis Crick when we articulated this program. Science should ultimately be able to nail this program. We should be able to understand like we’ve understand so many as other aspects of our life and the universe, science should be able to attack.
And solve and ultimately explain the problem of consciousness. And so we articulated this program just to take us back all previous cultures and even in our own culture here in the west until, roughly the 16th century, we thought the heart is the organ of consciousness, right? So you love with all your hearts, right?
You give your sweethearts heart-shaped chocolates for Valentine’s Day. Never brain shaped chocolates. But we do know, we have learned it’s a brain, but it’s not the entire brain that’s the organ of consciousness. For instance, you can lose your spinal cord and you’re quadriplegic, but you’re still conscious.
You can lose many bits and pieces of the brain, in fact, and you have all sorts of problems. But consciousness might not be impaired. And so the question is what are the actors, what are the cellular actors? We know the brain consists, like any other organs out of, gazillions of cells. In this case they’re called nerve cells or neurons.
What specific neurons in what specific location in the brain are necessary for me to hear or to see or to love or to fear?
CHAKRABARTI: Okay, and you thought that these questions would be answered in 25 years?
KOCH: Yeah. I was young. I just started a couple of years ago, I became a young assistant professor at Caltech, at the California Institute of Technology.
And you had believed in your methods in yourself. It’s just like you do a startup, right? Of course, you believe you’re going to make it. Of course, you’re going to believe you’re going to be hugely successful. Because this is what motivates you day in day out to do all the hard sciences and recruit students and look for funding in order to make progress, so you have to believe in yourself.
And I do believe, and I still believe we’ll solve this problem.
CHAKRABARTI: Again, I just think the rapid advance we made since Watson and Crick did describe that double helix, too.
CHAKRABARTI: From there to deciphering the human genome.
CHAKRABARTI: So your ambition and optimism, actually, is completely, was completely warranted at the time, but David. You made this bet with Christof, so were you similarly, I guess you weren’t as optimistic as he was about solving this problem?
CHALMERS: Yeah. I am fundamentally optimistic about a solution, but I think it’s gonna take a while.
The first thing I think to get clear on is that this bet on the neural correlates of consciousness didn’t resolve solving the whole problem of consciousness. Why do physical processes in the brain give you subjective experiences at all? Why doesn’t all this go on without consciousness?
That’s what in the business we call the hard problem of consciousness. Why is there consciousness? Why does it even exist in the first place? That I suspect even Christof thought would take longer than 25 years.
CHALMERS: This bet was about a slightly more limited topic, which is about correlations between the brain and consciousness.
Let’s take it for granted that consciousness exists. What does it go along with in the brain? What bits of the brain are active when you’re conscious? And I think both Christof and I thought this is a problem that could in principle be resolved even without solving the philosophical mind-body problem.
Without giving a full explanation of consciousness. The correlation project, what Christof and Francis Crick called the search for neural correlates of consciousness. I think we both saw her as a goal, which is visible somewhere in the distance, but maybe Christof saw it as closer. He was very optimistic and ambitious and thought maybe we could get there in 25 years.
My own view was, “The brain is so complicated, consciousness is so complicated that this question is going to be, it’s going to be a mess. It’s going to be complicated. It’s unlikely we’ll get there within 25 years.” If we did get there in 25 years, then great. But I was quite happy to put my money or my wine in this case on the other side of the bet.
CHAKRABARTI: Yes, can we talk a little bit about how you decided what was on the line, Christof, in terms of what would be won or lost in this bet?
KOCH: Yeah, no, we just agreed on a case of good wine. We did not specify the quality of the wine.
CHAKRABARTI: (LAUGHS) Okay. Okay so that was 1998. Suddenly it occurs to me, we’ll get to that hard question that you both talked about a little bit later in the show, but I had made an error earlier in saying 1988, and I thought to myself as you graciously offered the correction, that the hard problem is very hard indeed to solve, because one aspect of, at least, of my consciousness, is sensing that my interpretation of the passage of times changing that I have a hard time believing that 1998 was as far back as 25 years ago.
My mind wants it to be a much longer span of time. But so who won? Christof, who won the bet?
KOCH: So this event a couple of weeks ago in New York coincided with the release, the first press release of a large international consortium, of which Dave and I were also member.
Called an adversarial collaboration that was test, that was searching, questing for these neural correlates of consciousness by pitting two of the dominant theories of consciousness against each other. And these theories make different predictions about where in the brain these neural correlates are, and the timing of these neural correlates.
And this involves 12 labs and different continents that all agreed on a common protocol and functional brain imaging and EEG recording and magnetic field recordings, and 250 volunteers. To look at things and sometimes they see ’em and sometimes they don’t see ’em, or sometimes they pay attention to them, sometimes they don’t.
And thereby titrate out these footprints in the brain. And at this meeting, they announced the results. And so we took that as a great opportunity. Is there, if we look at these results, a unanimity, does everyone agrees this is the neural correlates of consciousness? And that was not the case at this meeting.
We disagree about the exact timing of it and exactly where they are. Clearly the field has not achieved, has not converged on a single answer.
CHAKRABARTI: Ah, so you had to give up the case of wine.
KOCH: So yes, I graciously agreed to Dave, and shook hand and handed him over a wooden box with six bottles of good wine.
CHAKRABARTI: Today, we’re joined by Christof Koch, he’s with us from Seattle. He’s with the Allen Institute for Brain Science. And chief scientist at Tiny Blue Dot Foundation. And David Chalmers joins us, as well. He is a professor of philosophy and neuroscience and co-director of New York University Center for Mind, Brain and Consciousness.
And we’re talking about how these two thinkers from the world of neuroscience and the world of philosophy came together to try, with a group of others, to begin to answer the question of where does consciousness lie in the brain? Now, David, can I ask you, I want to go back to that hard question. Because, when you said that things like when we feel pain, when we think about our mothers, when we see something move, that these may all be indicators of consciousness. Now I step back, and I think, okay in any scientific experiment we have to lay out our hypothesis and our presumptions.
And you said earlier that one big presumption that you were going to make was that consciousness exists. That seems to be a huge presumption, because these neural correlates about say, feeling pain. How can we say that those are correlates of consciousness versus simply just our brains processing physical information from the outside world and returning them as experiences that we’re having?
CHALMERS: Yeah. Consciousness existing, I think, is a kind of fundamental datum of our existence. We wake up. We experience the world. We see colors and shapes, we hear voices, we feel emotions. We think thoughts. All of these are directly experienced. René Descartes, back in the day, said, “I think therefore I am.”
And I think he really meant, “I am conscious, therefore I am.” Consciousness is the one thing we’re sure about. I think what we’re not sure about is how it connects to processes in the brain, and how it is that processes in the brain actually give you subjective experience in the first place. That’s what we call the hard problem of consciousness.
Actually, Christof and I first met at a conference in 1994 in an early scientific conference on consciousness where he was talking about the project of neural correlates of consciousness, and I said, “Okay, that’s well and good. But correlation doesn’t necessarily give us explanation.” Yeah, merely having a story about these processes in the brain might help to tell us how we respond to stimuli and so on, but it doesn’t really answer the fundamental question of, “Why are we conscious at all in the first place? How do physical processes in the brain somehow give rise to this magical experience of consciousness?”
CHAKRABARTI: Okay. So maybe I should just ask the fundamental question of both of you and David, I’m going to stick with you for a moment. From your perspective as a philosopher who’s also knowledgeable about neuroscience.
We keep referring to Descartes about, “I think, therefore I am.” But how would you define what consciousness actually is?
CHALMERS: Yeah, it’s not easy, but there are a couple of simple definitions I like. The most simple definition is simply subjective experience. Any experience that feels like something from the inside.
So something as simple as seeing a red apple in front of you, you have a subjective experience of this. If you like, inside the inner movie of the mind, you hear your own voice. That’s part of the inner movie of the mind. It’s a multi-track movie. You feel pain, you feel emotions. You think about your hometown.
All of these are states of subjective experience. There’s something, it’s like, there’s something it feels like for us to go through these states, and it’s that subjective experience, which I think is really at the heart of consciousness. That’s certainly the most mysterious aspect of consciousness, the most familiar thing in the world, but also the most mysterious.
CHAKRABARTI: Okay. So Christof then, from your perspective in neuroscience, how would you define what consciousness is?
KOCH: As Dave did, it feels like something to be in love. It feels like something to hear these voices right now in my head. Look, it’s the only way you know that you exist, Meghna. If you go in a deep sleep tonight, you’re going to go to sleep.
And at certain parts of the night, you’ll be in a deep sleep. We can see that because your brain will, there will be these slow waves crisscrossing your brain, okay? At that point, you do not exist for yourself. Meghna is not there. Meghna doesn’t exist for yourself. Of course, your body still exists. Your brain still exists, but you don’t exist for yourself. When you’re in a coma, right, some patients are in a coma or when you are anesthetized, it doesn’t feel like anything anymore. So that’s the difference. It’s a difference between life, between having any experience, having any sensation and having no sensation. And the challenge has been that if you look at the foundational equation of physics, so you look at general relativity or quantum mechanics, there’s no consciousness there.
You look at the periodic table of the elements, there’s no consciousness there. You look at their genes, there’s no consciousness there. Yet here we wake up. Here we are in a world having, seeing and hearing and loving and fearing. So how does consciousness get there? So that’s what Dave calls a hard problem.
Whether it’ll always remains hard is an open question, but surely, it’s a conceptually very difficult problem. So that’s why we focus on the easier problem. Where are the footprints of this, where are the correlates, the bits and pieces that correlates with me right now feeling in love and 10 seconds later realizing that I’m late for an appointment. Where are they?
CHAKRABARTI: Again, we’ll discuss the neural correlates more, but I suppose it’s inevitable that I’m just intrinsically fascinated by the big question of what is consciousness? Because Christof, when you mentioned in sleep, I’m not aware of myself.
KOCH: In deep sleep.
CHAKRABARTI: In deep sleep. Okay. So not REM sleep or dream state, because dream state —
KOCH: That’s correct.
CHAKRABARTI: Okay. Go ahead.
KOCH: Entirely correct. In dream state, your consciousness, it’s not the same waking consciousness. So for example, typically you don’t realize you’re dreaming, right?
Typically, yourself is muted. It’s more like you’re in this movie and things are happening, but you’re typically not the actor or the director of your dream, right? That happens in these dreams called lucid dreams. So dreaming is clearly a state of consciousness, but then there are many states in between when you’re not conscious, when you’re in deep sleep.
CHAKRABARTI: I see. Okay, good. Because I just wanted to check that because sometimes my most vivid subjective experiences or a person’s most vivid subjective experiences come exactly in that dream state.
KOCH: Exactly. And you can’t tell the difference between dreaming and real life. It feels like that’s what life is. It’s experiences.
CHAKRABARTI: Yeah. Yeah. So David, taking another look at the idea of this, of consciousness being the subjective experience. Would that mean that consciousness, the experience of consciousness, lies well beyond just the human realm. That any manner of animals might be also similarly conscious?
CHALMERS: I think the consciousness goes well beyond the human realm. This is a matter of controversy. Descartes, René Descartes, back in the 17th century thought that consciousness was for humans only, and that even his dog was not conscious. These days, I think there’s a very strong scientific consensus that at least many non-human animals are conscious.
I think, most people would say that most, probably all mammals are conscious, probably all birds. Now there’s a debate about fish. Conscious? Are insects conscious? None of this is to say that they would think or even that they’re self-conscious.
Self-consciousness is a lot more, is a lot more complex. Maybe that’s humans and some other animals, but basic subjective experience of the world, like feeling pain, like seeing colors. I think most scientists and philosophers these days think there’s a fair amount of evidence that at least a number of animals have subjective experience.
And then of course we can go on to ask questions. Does it go beyond that? Could a artificial intelligence system, for example, have subjective experience? And these questions are becoming very relevant right now.
CHAKRABARTI: I think this is a problem of measurement, right? We can’t fully know what a, I mean, there’s been treaties written on this about like, how does a bat actually feel? Because we can’t know the experience of a bat, right? There’s no way, there’s no way for us to measure it. Maybe some day there will be.
CHALMERS: In philosophy, this gets called the problem of other minds. How can you ever know that any other system has a mind or is conscious?
I’m totally confident that I am conscious. Can I know for sure about you, Meghna? Or you, Christof? I believe you’re conscious because you’re like me and I’m conscious, but we don’t know for sure. Once you go to animals or machines, this problem gets all, gets that much harder. I’ve often thought it’d be so much easier if we had a consciousness meter.
We could just point at a system to see, to measure its consciousness. But that’s precisely what we don’t have, because consciousness is private and subjective.
KOCH: But some people are trying to develop such a conscious meter.
CHALMERS: It’s true with the health of neuroscience. Once we know the neural correlates of consciousness, maybe we could use those neural correlates as a guide to consciousness.
On the other hand, the fact that consciousness is hard to measure is one of the things that makes it quite difficult to find the neural correlates of consciousness in the first place.
CHAKRABARTI: Ah-huh. Interesting. Now, David, when you said you believe that I am conscious. I’m in Oregon right now.
You’re in New York State. Christof is in Washington State, so we can’t actually see each other. And David, your belief that I am conscious rests on the trust that you have in my, on the On Point team who told you that there’s this person named Meghna who’s going to be hosting the program. Even though you can’t see me, all you can do is hear me, which just got me.
You talked about machines, right? We’re rapidly entering the stage of human evolution where a machine may be talking to you but it’s getting harder and harder to distinguish whether or not that’s a conscious human. Or an artificial intelligence. It’s a little disconcerting, isn’t it?
CHALMERS: Yeah. I’ve only heard your voice come to think about it and I’ve read about you and so on. But yeah. For all I know, this is actually a large language model. Chat GPT in the crowd who they say, “Oh, Meghna wasn’t available today. Let’s get a language model trained on everything Meghna has said, and the language model will probably be able to fake it enough.”
I think the technology may not quite be there for that, but in probably in three or four years it might be. And that really raises the question. Now, Christof, I’ve actually seen in the flesh and talked to and so on. So I have a bit more evidence of that. But who’s to say maybe some false memories got implanted.
Maybe all this is a giant virtual reality, maybe I’m the only conscious being in existence. These are all philosophical thrillers to think about. The science tends to set them aside by assuming that at the very least, all humans are conscious. If you really wanted to raise trouble, Descartes would probably say, “How do you know?”
CHAKRABARTI: Yeah. As far as I can tell, we have not yet entered the matrix, and I can guarantee everyone that I am a living flesh and blood human and not a bot just yet. But Christof, could you describe, I read about some recent, relatively recent experiments that you and a team that you brought together called the Cogitate Consortium did in order to start collecting some data to answer this question about the neural correlates of consciousness. Can you describe some of the experiments you did?
KOCH: Sure. They’re relatively straightforward, so you have volunteers that lie either inside a magnetic scanner, so we can pick up your brain’s hemodynamic response, the bits and pieces of your brain that are most active because we can track that in the blood flow of those regions of the brain.
Or we had volunteers where EEG nets so we can track the electrical activity. Or they were inside an MEG device that tracks the magnetic fields associated with thinking, overseeing. Or in some cases these were patients that had volunteered for these experiments. And these patients had electrodes implanted into their heads to monitor epileptic seizures.
If you have certain types of epileptic seizures in order to track where they originate. And can you maybe surgically remove them. They have these electrodes implanted. And then we had all these people look at simple stimuli. Face that was clearly present or an object, like a toaster or a telephone or letters or some other simple symbols.
And sometimes they had to push a button, and they were asked to push a button when a particular face appeared or a particular or object appeared. But not when there were other distracting faces. And so this is a fairly standard experiment. What makes this different is a very large number of subjects involved.
250. Most experiments have 10 or 12 subjects, and then people come to some conclusion based on 10 or 12 subjects. We call those often statistically under power. They don’t have enough statistical power in order to really come to a conclusion with high confidence. The other thing that this involved, all these different techniques doing very similar protocol.
So that we can compare the EEG, the electrical response with a magnetic response, with a response inside the magnetic scanner. So that was the nature of this experiment. The other thing that was unique about it, it involved two, as I mentioned, these two dominant theories of conscious, scientific theories of consciousness.
On the one hand, there’s integrated information theory of consciousness, abbreviated often as IIT, integrated information. On the other hand, there’s a theory called Global Neuronal Workspace Theory. They’re quite different. IIT stresses that consciousness is very rich, that if you look at the world, there’s sort of conscious content everywhere.
While Global Neuronal Workspace Theory stresses it’s more related to thinking and making a high-level summary. It’s more cognitive theory. So it’s a little bit like IIT says the world is like a painting of Pieter Bruegel, where you see these rich peasant scenes and you can see myriad of details, the way they’re dressed, the way the people are looking at each other.
While Global Neuronal Workspace Theory says, consciousness is more like the label you might see at a museum. Flemish painters, 16th century peasant wedding people, making out and drinking. And these theories came to different conclusions about where the NC is. IIT says it’s in the back part of the cerebral cortex, the outermost layer of the brain, while Global Neuronal Workspace says no, it’s really in the front of the brain, the front of the brain that’s most expanded in us compared to other animals and so that’s one way these series distinguish themselves.
Is it in the back or is it in the front? And then there are some other differences about the timing of these signals with respect to how long, when I’m conscious of something for 10 seconds, one theory says there should be a footprint for 10 seconds. The other theory says, “No. It’s really only early on when you first become conscious that you send out this broadcast to everyone in the brain.” And you only are going to pick up this broadcast at the beginning of seeing something. Or when it disappears at the end. So this is how we try to distinguish the two theories in terms of their predictions.
CHAKRABARTI: Today, Christof Koch is with us. He’s a scientist at the Allen Institute for Brain Science and chief scientist at Tiny Blue Dot Foundation. He’s in Seattle, Washington. And David Chalmers joins us. He is professor of philosophy and neuroscience and co-director of New York University Center for Mind, Brain and Consciousness. And he’s joining us from Yorktown Heights, New York.
And David, I would say that one of the greatest delights of science is that well-designed experiments asking intelligent questions only beget more questions. So the endless pursuit of knowledge continues. And listening to how Christof was describing the experiments in the previous segment, about seeing like where would the signals of consciousness come from, the prefrontal cortex in the front, or instead in the back of the brain. And it looks like the experiment revealed it may be, the results were inconclusive, right? It made me wonder, I don’t ever want to say any scientific pursuit is a folly, but is this pursuit one that’s destined to never come up with a clear answer?
Because we’re coming at this from the perspective that science being objective can find an answer to the question of consciousness, but we’ve also defined consciousness as a subjective experience. So how can the twain ever meet?
CHALMERS: Yeah, I am an enthusiast about the science of consciousness. I do think it’s something, consciousness is something we can study scientifically, but it does pose many unique challenges. One of them is the fact that it’s difficult to measure directly in other systems. We know about consciousness in ourselves, but how do we measure it in another system?
Yeah. The fact that it’s so subjective, by its nature. But that said, I think, we tend in the science of consciousness to at least adopt the working assumption that, for example, when someone says they’re having a conscious experience, then they are conscious of what they say they’re conscious of, at least if there’s no reason to say otherwise.
And that allows us to leverage verbal reports. We can do experiments where we ask people what they see and use that as a guide to what they’re conscious of. But still many philosophical issues arise, and this is an area where scientific and philosophical issues are entangled all the way through.
I think when Christof and Francis first started approaching this in 1990, around 1990, they said, “Now we can cut through all of that philosophy and just do the science.” I think what we’re now discovering, even with some of the results of these adversarial collaborations, is that there’s a lot of philosophical assumptions intertwined with your scientific results.
For example, the difference between Integrated Information Theory and Global Workspace Theory, which were tested. These experiments is, I think, partly a philosophical difference between two different conceptions of consciousness. One where consciousness is very rich, very sensory, like a picture, as Christof was saying, and the other where consciousness is very sparse, much more tied to thinking than tied to sensation and so on.
And it turns out to be quite difficult to test this philosophical difference experimentally. So, one thing I’d like about these adversarial collaborations, the experiments that were done, is that each of the theorists was forced in advance to make certain predictions about what the results would be. And to say if this prediction doesn’t come out, then my theory is potentially falsified, or at least challenged.
One thing we actually, it’s interesting now we’ve actually seen the results. You might say, “Okay, we’re going to get some results.” Some theorists, on both sides, predictions were, in fact, challenged. They said we would find certain activity in prefrontal cortex, and we didn’t find that. What we, interestingly though, what we now find is that the theorists say, the Global Workspace Theorists, “Oh, maybe we weren’t actually conscious of that stimulus after all.”
Yes, it was in front of you, but maybe people weren’t paying attention to it, so maybe they didn’t see it consciously. And to some extent, that reflects this philosophical difference between two very different approaches to consciousness. One where it’s very rich, one where it’s very sparse. So even interpreting the scientific results, I think, takes some philosophical thinking.
CHAKRABARTI: Oh, that’s interesting. That’s a twist. Christof, then, can experiments around consciousness ever be scientifically conclusive? Because A) once a hypothesis has been tested and either proven or rejected, it has to be reproducible in order for science to accept it. But if David’s saying that even the interpretation of the experiments results are subjective to philosophical considerations, can those results be reproducible in the conventional scientific sense?
KOCH: Yes. So firstly, let’s not be defeatist.
KOCH: There have been many problems in the history of science that, where people argued with good reason at the time that they would never be solved, right? Like we will never know what stars are made out of. We know what stars are made out of. We shall never understand life.
We shall never understand the difference between life living and non-living matter. We understand that, right? So there are many things that people previously had argued for good reasons we couldn’t solve. We have understood. Alright, so let’s not sort assume that consciousness somehow is beyond the bane of human ingenuity.
CHALMERS: And I should say, I totally agree with Christof about this. I think the problems are hard, but we can solve them eventually.
KOCH: Good. Okay, so we got that. B) they are reproducible, in the sense that if we now take 250 different people, okay, we do the same experiment. The chances are very high, 99% plus that they will find this, that they will have the same finding.
That would still leave open the question that Dave just state, “How do we know in this particular instance that they were conscious or not conscious?” But these experiments are reproducible, and they will also have, Meghna, a lot of practice and tracking down these colleges, not just some esoteric pursuit of philosophers or science geek, it will have practical problem, practical consequences like in patients.
There are thousands of patients today on the planet in ICUs or in homes where we don’t know whether they’re conscious or not. Terri Schiavo may be a name that you might recall from 20 years ago.
CHAKRABARTI: Mm, yes.
KOCH: When these patients are in a persistent vegetive state and probably 20% of these patients, people estimate, actually do, are conscious, but they’re unable to signal it because they’re so severely injured.
Okay. If we have the neural correlates of conscious, we could build this conscious meter that people are trying to do and say, “Okay, this patient is actually conscious. They’re inside their brain. Someone is there compared to most of these patients where there’s no one home anymore.”
Once we have it, we can track it. When does it first begin? Is it in the second trimester? Some people argue of a fetus, or does consciousness start when you’re just born, right? We can track it on the assumption that it’s similar in us. Between us and let’s see, other mammals, because the brains are very similar.
We can track it in a dog, or in a monkey, or a great ape or in orcas, right? So there will be lots and lots of practical consequences to finding the neural correlates of consciousness. And thereby we can partially validate these experimental findings that are more scientific. We can validate them in clinical, in the clinic.
CHAKRABARTI: Yeah, so can I just jump in here for a second, Christof? Because first of all, I want to agree with both of you that I don’t prescribe to scientific defeatism either. But perhaps it’s just in my layperson’s mind who’s stumbling repeatedly on the same obstacle, and that is, okay, I hear you when you say that the results of these experiments are indeed reproducible.
But you also then, I think, your next sentence was, “It’s just that there’s a difference of opinion about whether in that particular moment the person was conscious or not.” But if there’s a difference of opinion in that really fundamental state that you’re even trying to measure, how can we, even if the results are reproducible, how can we assign any meaning to them?
KOCH: More experiment.
KOCH: It’s a good question. So the question that was posed by these theoreticians, right? How do we know in this particular trial that the patient, that the subject is conscious? We can think of another experimental test, right? So all of these questions, there will be some way we can test these questions, right?
And so ultimately —
CHALMERS: I think what we’re thinking about too is that there may be easier aspects of consciousness to study, and harder aspects of consciousness. So for example, consciousness within attention, when you’re attending to the stimulus and you can talk about it, that’s a case where I think most of us can agree that, “Okay, a human subject will be conscious of the stimulus on that occasion.”
These harder cases that we’re arguing about now is like, “What about a stimulus that’s in front of the subject, but they’re not attending to it?” Some people think that could still be conscious, some people think not. That’s hard to get at experimentally, but I think at the same time, we can all focus on the easier cases like consciousness within attention, things you can look at and talk about right now, and so on.
And at the very least, try and form a science of those aspects of consciousness.
CHAKRABARTI: So that makes a lot of sense. Now I appreciate both of you guiding me through that. So at right now we’re in a phase where if we can find sort of subsets of consciousness around which there’s both scientific, and if I can put it this way, philosophical consensus about what’s actually happening, then we can pursue experiments in those subsets.
But, I don’t know. Maybe I just keep getting dragged back to the big questions. Because David, it’s impossible not to have the mind wander to, again, how do we define consciousness? That some people might say aren’t you simply talking about the soul? Or am I bringing too much philosophy or religion into the question?
CHALMERS: Yeah. I would prefer to resist equating consciousness with the soul. Because the soul, who knows what the soul is, but it comes with many religious connotations. The moment we start talking about the soul, it connects to life after death. For example, the soul is something which might survive our bodily death, maybe created by a creator.
Maybe it gets reincarnated. It’s a wholly non-physical, something totally separate from the brain. I’d be inclined to resist most of those connotations. I don’t know that consciousness survives death. Maybe there are non-physical aspects of consciousness. But I think it’s going to be, at the very least, very closely tied to the brain.
So I preferred, I think subjective experience for me is a datum that we’re conscious. The soul to me is a theory. Maybe we have souls that survive our death. Maybe we don’t. I don’t know that question. So I think of consciousness basically as a scientific fact. But still, it is an incredibly challenging scientific fact. Because consciousness seems so different from everything in the physical world.
We’ve got this program of trying to explain everything in physical terms. Physics explains chemistry, which explains biology and so on. And then suddenly consciousness seems like somehow an exception to that explanatory order. Our old methods of trying to explain things like say this neuron fires and that causes this behavior, just leaves open the question, “How does that give you consciousness?”
So I’d like to think it’s a hard problem, but still, that doesn’t mean we can’t solve it scientifically, it just means it’s a new kind of science, potentially.
CHAKRABARTI: But yeah. So it does sound like you’re saying though that you’re encouraging us to think about consciousness as a fundamental part of the universe in addition to space, time, mass.
KOCH: Yes. It’s a feature. Yes. It’s a feature of the universe that we find ourselves in.
CHALMERS: This is a radical view that I think.
CHAKRABARTI: Tell me more.
CHALMERS: I’ve always been somewhat sympathetic with. I think Christof has become more sympathetic over the years. That maybe, in science, you have to take some things as fundamental space, time, mass, charge.
In many basic scientific theories, you don’t explain them. They’re just fundamental ingredients of the world, and you give theories of the laws that involve them. So I’m inclined to think consciousness could end up being fundamental. In the way that space, time, mass and charge are. That doesn’t mean you can’t have a theory of them, you can still have a theory of space and time, even though they’re fundamental.
It’s just our theories will need to describe the structure of consciousness and articulate the laws that connect to physical processing. So if, for example, consciousness depends on certain kinds of information processing that happens in brains, we need to articulate the fundamental laws that say, “When you get this kind of information processing, you get this kind of consciousness.”
We’re not there yet, there’s a research project underway.
KOCH: So this is exactly what one of those two theories, Integrated Information Theory, it starts with consciousness. So it assumes that consciousness is a fundamental feature of the universe and of mechanisms like brains or like other physical mechanisms.
It says that any mechanism that has complex interaction will feel like something.
CHAKRABARTI: Any mechanism that has a complex interaction will feel like something.
KOCH: The theory says that any mechanism that, technically speaking, that has intrinsic causal powers upon itself, yes, will feel like something.
So share some intuition with this ancient philosophical or religious belief called panpsychism, right? That everything is in soul. Maybe not everything, maybe not tables and chairs, but many more systems that have complex internal interaction feel like something. That’s what, at least the implication of Integrated Information Theory is, including big brain animals.
They feel like a lot. Like you and I, in fact, they can feel so much they can begin to reflect upon themselves and then thereby be self-conscious, which creatures with smaller brains probably don’t have, like a mouse. There isn’t really too much evidence that the mouse knows it knows something.
But the theory also says maybe even simple, simpler system that don’t even have a nervous system, right? Potentially plants, potentially single cells, because even a single cell has vast complexity that we’ve never managed to model on a computer using computer simulation, because these complexities exceeds anything we can model today. So it may well also feel an itsy bitsy bit like to be a little amoeba or a little bacteria.
CHAKRABARTI: Ah, interesting. David, we have just a couple of seconds left. I understand you and Christof have entered a new bet. Do you hope to win again?
CHALMERS: Yeah, with the next bet, another 25 years, Christof bet that will have discovered the neural correlates of consciousness by 2048.
And I’m still taking the no side. I think we’ll get there eventually. That’s still optimistic, but I would love it if I’m proved wrong, if we find [that] neural correlative consciousness, fantastic. I still think I’m on the winning side, but let’s see.
This article was originally published on WBUR.org.
Copyright 2023 NPR. To see more, visit https://www.npr.org.