Transcript of journalist and senior media executive Richard Sergay's interview with Lucia Melloni, PhD for the "Stories of Impact" series.
RS = Richard Sergay (interviewer)
LM = Lucia Melloni, PhD (interviewee)
Lucia Melloni is a research group lead at the Max Planck for Empirical Aesthetics in Frankfurt, as well as a professor at NYU Department of Neurology. This is a transcript of her conversation with interviewer Richard Sergay, for his documentary series "Stories of Impact."
RS: Your background and interest in the issue of consciousness?
LM: So I am a psychologist for training. I was trained first as a clinical psychologist. I was fascinated by Freud on the unconscious. And the unconscious brought me into neuroscience and into consciousness. It was there I did the opposite, travel. And it’s going to become interesting as we speak, I can tell you, because I think that I am a very good example of a lot of the things that people face usually, which is we take consciousness for granted and we think that the unconscious is what is the mysterious part. So I did exactly that trip. So I was trying to (INAUDIBLE) psychologist first and then since he was fascinated by this idea of the unconscious desires and this, there’s somebody inside me that rules me but I don’t know that person. I became interested in what were the underpinnings of that. So what is the difference between me being aware of that, versus me not being aware of that. And then ended up PhD in neuroscience. Actually trying to find out-- by the time I had my PhD in neuroscience, I realized that the mystery was actually the opposite was consciousness, not the unconscious. And then in that PhD neuroscience, working on masking and, and what the neural correlates of that experience could be. And we’re working at the time on the hypothesis that long range synchronization. So the idea that you know collections of neurons across different areas of the brain would be the correlate, and we provided some good and nice evidence for that. And then I’ve been working much more on trying to understand how we can expand our consciousness capacity. So instead of taking consciousness for, ok, this is what it is. Is there a way in which he can make it, in which he can enlarge it. If I train people, if I do meditation for instance. Etc. So is it, is it malleable or not, with fixed. So that’s more or less my background.
RS: What is consciousness?
LM: It’s in many ways, you know, I think the best way in which I, I always tried to portray it is, is this experience. So, you know, you’re drinking a coffee and you’re enjoying your coffee...
Experience. So, you know, what makes us enjoy, what makes us taste, what makes us, you know, have feelings, what makes us, you know, be sad. And is that-- and what, and one important aspect of it, it’s what makes it pretty sometimes difficult to investigate, is that it’s private. So I know that I have these experiences, but I can’t be sure that you have those experiences, right. And they can tell you very clearly when I am having those experiences versus when I’m not having those experiences. So for instance I am a lucid dreamer and, you know, I can tell you in the middle of the night I wake up and I can control my dreams. I'm there experiencing them, but of course if you look at me from the outside, I am immobile. So in many ways it seems to you as if I’m not there, right. And that’s the, that’s the challenge from the scientific point of view, but in many ways it’s, you know, it’s simple because we know that it exists and we know very, very well, you know, when if I am under anesthesia it’s not there. If I am under deep sleep it’s not there. If I have a trauma in my head, I lost it. And if I have epilepsy, I happen to work a lot with patients with epilepsy, they can tell you like I wasn’t there and they wouldn’t remember anything, so it’s very clear for us from the internal perspective when we have it and when we don’t.
RS: Where is consciousness in the human body?
LM: I mean that’s what we’re trying to find out, right. So of course we locate in certain parts and not in others, so we say ok, it’s not in my liver, it’s only my stomach. It’s not in my eyes. We think this in the brain. But the whole question is where in the brain and why in the brain, right. So, and this is what we think now, right. I mean I’m sure that you know, I mean many years ago we said it was in the pineal gland. Before in the heart. I mean theories have been evolving, right. The current understanding is that the brain is the most likely candidate. And that’s because we know, and this is something that is interesting because that, we know it a lot from patients that have lost it or have lost aspects of it, right. So we know that if you, if you have a tumor on that part of your brain you know has to be resected, then you sometimes lose a function. So you cannot see colors or you cannot see movement. Or you cannot see faces, right. That is very specific and that’s how we know that it seems that the brain is the part that relates to consciousness. But more than that we would want to know why, right. So, and which parts. Is it the whole brain, is it a little bit of the brain. Is it you know, and what, what in the brain, is it the connections of the neurons. Are the neurons themselves, is it the way in which they are arranged. Do they have to be hierarchical? Do they have to be interconnected? You know, and these are the questions that we’re trying to address.
RS: Where are we in understanding the brain, the mind, and consciousness?
LM: I mean we know, you know, at what point different, we know a whole lot more now than we knew say 60, 70 years ago, but this doesn’t mean that we know a lot, right. So it’s all relative in some way. So now we know like how neurons communicate, you know, we know that certain neurons encode for instance orientations with the advent of, for instance, invasive techniques, like FMRI, for instance, that we can look into the brain. So we have, there’s many, in many ways we have advanced significantly, right. Now it can put you for instance in this counter and it can ask you to think about, you know, like a place where you have been and I know which part of the brain will light up. This, this we didn’t know many years ago. Now, is that understanding of the brain. I don’t think so. It just tells you that ok, there’s a part that’s just a map, right. But you know would you, would you know how to navigate in New York City just because you know, you know like, there were the streets. No. You need much more than that, right. So in many ways we know more. But in particular for the human brain is hard because we don’t have access directly. We always have access through the measures such as having MRI for instance, which give you a very nice picture of the brain. But it’s not necessarily new activity. They measure something slightly related but it’s not, it’s not neural activity. And on top of that they don’t give you the time, they don’t give you the timescale in which (INAUDIBLE), right. And then there’s other techniques like for instance, EG, that have been around for many-- for, for you know, quite, quite some years, I think over a hundred years or so. But they don’t give you information about where things happen, right. And then you can take a lot of advancement in animal models and we know, we know a lot about animal models. But then again, there is this challenge all the time that, you know, I’m not a mouse, I’m not a monkey, I’m not a, I’m not, you know. And so there’s always this question like ok, how much of what we know from other animals translate or doesn’t translate into humans. And the-- and, you know, doing cross species comparisons is not easy, right. So we know much more. There is many, there’s many new techniques now that allow us to you know, trace neurons, you know, you know, especially record large portions of the brain, you know many, many years ago people would just record with one single electrode in one area and then of course, you might want to understand something that is very big with you know just one microphone. Like (INAUDIBLE) here. You want to capture like this, the conversation in the stadium with one microphone, of course you know you don’t understand that, right. But still is incomplete, right? And the same thing happens, you know, the same thing holds true for consciousness, but for any other, like if you want to understand memory, we’re facing the same challenge. It’s not special, specific to consciousness. Our techniques are limited. And in, you know, when we’re missing a good telescope it’s just hard to see the stars.
RS: What are the main challenges now in trying to understand the brain? You face enormous challenges?
LM: Well, so the technology is something that we should not underestimate. So we need better technologies, right. And I think a lot of-- new research initiatives are in developing technologies, in developing causal manipulations that allow you to you know manipulate neurons specifically, right. So without those, you know especially for the human brain, right, as I said that you know, keep in mind that all of the technologies that we have for the human brain are all and all these non-invasive ones. Would give you access to a specific neurons, right. And we have billions of them and we have billions of connections. And without mapping those, you know, it’s the understanding will be limited, right. Then, in light of limited instrument, then the evidence that we have is, you know, to say the least incomplete, right. And that also leads to theories that cannot be, you know, complete either, right. So there is always this exchange between if I have good data and if I have good understanding then of course my theories would be better, right. And this is, at this point, you know, we just need to do more incremental science and I think people have realized this challenge. And they have realized now more than ever, then, that part of, you know, facing the challenge will require, you know, consortia. It’s not like, scientists used to do science by individual labs. I would do one thing and I would not team up with somebody else and I think that people outside realize that this is possible. But, you know, we will advance science in a very limited way and it would take us much longer. Whereas if we actually collaborate, and you know this pandemic for instance, COVID is a beautiful example of what are the things that you know people have done now, they realize that of course you can have many, many different labs trying to find the vaccine or they can actually try to share data, try the different compounds and then see what happens, right. And the idea is you would probably advance, you know, further with that, right. So we need, there are many different reasons why we are not as far as we would want to. This doesn’t mean that, you know, we don’t know anything, of course we know some stuff, people do, you know, when somebody has epilepsy for instance, you do undergo a surgery and then the they extract a part of your brain and then we know that if you extract these part of your brain you might lose, you know, language, you might lose memory, so there is something that we do know, right. It’s not complete darkness.
RS: Link that to understanding consciousness?
LM: Well so for instance, we know that, there is there’s many different ways of understanding that question, we’re answering that question. So on the one hand we know from lesioning studies as well as from FMRI, and the evidence has been consistent in those cases, that for instance, whenever you see color there is an area of your brain that light up, and if that area is, it has an insult, because of you know traumatic brain injury, or etc, then you lose that capacity. Right, so then we know that there are certain parts of the brain that are related to specific, because when we talk about consciousness, it’s you know, we need to say ok, conscience is about something. It’s about colors, it’s about movement, it’s about you know hearing, and so on and so forth, right. So then the properties of that experience, we know that are linked to specific areas of the brain, right. So now the question, and this, and I think for that we actually have very good evidence and even more so. So for instance if you take the very good example of face processing. There is an area in the brain that people are it is called the fusiform face area. This was actually found by FMRI many years ago by you know Nancy Kanwisher, then researchers doing monkey physiology, what they did is they put the monkeys on this counter. They did FMRI, then you know, they found all those areas and they put electrodes, and they found out that those areas in particular had neurons that would respond to particular faces. Right. And furthermore, like if you stimulate those areas, very often you would also have illusions, right, because the whole point of consciousness is like, if this area is really encoding that property, and if I inject activity in that area, you should also hallucinate that, think of, you know, what happened when we, when we dream. Of course is not coming from the outside, right. I can still see faces and experience colors, etc, when I’m dreaming. So this means that somehow this is in my head, it’s not necessarily outside. Of course it isn’t in tune with what is outside, but I can also produce it inside, right. These are all the things that we do to prove that there is certain areas where information is more relevant than orders. And also, we know for instance from a lot also from patients with neurological disorders, that there are certain areas, and if you ask, if you talk to neurosurgeons they would say look, there are certain areas that we cannot touch. And you know that’s why we do functional mapping, so functional mapping means that you know like, if you stimulate an area, and if a patient, you know you ask them to talk, right. And as they are talking you stimulate that area, and if they cannot talk anymore you say ok, look, that’s the, that’s the seat of speaking, right. So then, you know, you have to preserve that area. But then there’s a lot of other areas that appear to be silent. And those are the ones that you know neurosurgeons resect, because they think that they actually do nothing, you know, for some functions or consciousness etc, right. There is for instance also interest, and you we’re asking for what do we know about consciousness. You know. Which areas are really important. Also from electrical stimulation in epilepsy patients, we know that the, there’s parts that you stimulate and people may lose consciousness, right, or the other way around, people who are in any sort of consciousness, if you stimulate an area, they might kind of like awake. So we know that there are certain areas that seem to be more relevant than others. Now are those this seat of consciousness or are they part of a network. And that’s generally where a lot of the disputes and a lot of the questions actually remain, right.
RS: So you’re talking about having lucid dreams. How you define consciousness in the dream world vs. the real world?
LM: To answer that question-- so from my internal perspective, I can definitely tell you I can definitely make that distinction, so I know when I am sleeping and when I’m lucid dreaming, right. Not just because the dreams are irrational in certain ways because by controlling them you can make them more rational, right. But you know that, you know you, have physical feedback, so you cannot move. So then you know that you are like locked, right. So from the internal perspective you can make that distinction, right. And I think when we dream, we know that this is not happening for real, right. That’s, you know, just from the you know, from the non-scientific point of view. Now, the community has defined those as the distinction between states of consciousness, right. And then the idea is that, in certain ways you can think that you have different states of your brain, right. But that nonetheless, in certain states, you can have the same conscious capacities right. And we know that for instance when we are dream-- when we’re sleeping, there is a part of that state of sleeping by which we don’t have consciousness, right. So when I when I say I’m a lucid dreamer, I am only lucid dreaming doing this bit of REM. In which I kind of like wake up. But when I’m in non REM sleep, I am definitely not there. And is it pitch black. And pitch black in the sense that I don’t know, you know, like I’m not there, I am not perceiving anything, I’m not having any thoughts, you know, there’s just nothing. And we know that that exists, right. And that part, you know, what we know is that our brain in that state is not the same as our brain during the awake state. Right. The waves in our brain are different and we can, you know, we may look at the waves of the brain from the REM state and the brain waves during the awake state, they’re actually very similar, despite the fact that of course you know, with the, whether we move or not is very different. Right. But now if you take the brain-- the brainwaves of the non-REM sleep, they are very different from the ones of the awake, and from the ones of the REM sleep. So we can distinguish those. And also based on behavioral measure, right. So it’s actually, you can, you can see that you know, somebody when they are sleeping, the muscle tone is completely gone. Right. But then the brain wakes up, right.
RS: Where are we in 2020 in understanding of the arc of understanding the brain?
LM: Well, so, we do know a lot of, you know. So there is always a lot to discover for sure. But I am hopeful and I think that we are in a far better place than where we were, you know, 50 years ago. So the knowledge is higher better. Significantly better than you know, 50 years ago, especially in humans. We can do much more than we used to do before, so then we are, you know, so now for instance, I work with patients that have epilepsy. They have opened our eyes in terms of now we can have electrodes inserted in people’s brain and we can have a much better understanding of the brain made by having recordings from within directly. These are things that they were possible before, but the technology wasn’t necessarily there. So there is-- there is a significant progress there. FMRI is also getting much better. So now we can, you know, we have FMRI at 7 Tesla, you can do laminar or you can infer what the delay distribution of the different areas you are doing. So how the information gets in and how the information goes out of an area. So there is, there is a lot of progress. This doesn’t mean that we know everything. This doesn’t mean you know, we have cracked the mysteries No. But we are in a good way and people have also realized that the technologies will play a significant role, right. So the brain initiative for instance is a lot for development technologies, right. And because they recognize that without technologies, our understanding will be limited. It’s not, you know, it’s not only about being clever and running clever experiments if you don’t have the proper telescope or the proper instrument. You know, you just can’t see, right. So it’s very simple. And we have, and those, those have been developed. Or under development, right. So I think there is a lot of progress there.
RS: The biggest questions that need to be answered about consciousness, the top five?
LM: Well, it would just have one, I would say, not even top five. But there’s ones like, you know, why are we conscious. You know, why matter causes consciousness, you know. And this is, you know, understanding it in the brain, in physical systems. You know, once we know that we will know the basics, and then you can ask, and then, why does it feel red. And why does it feel the green, and not yellow, right. But you know just having that seem, answering that question would be, you know, is at the core of what the problem is, right. Because the-- the issue with consciousness is that we don’t, or we take the world as physical reality, right, as particles. And you know, the question then becomes, what in those particles or in whatever constellation of neurons or molecules etc., cause-- you know, produce consciousness, right. And you know, we think that it is in the brain because you know, that’s evidence that we have now. But this also doesn’t entirely entail that brains are the only, are the only organ, then can, that can host consciousness, right. And that would be at the wrong, so the-- the conclusion is that brains can-- and are they the only, no. Or we don’t know the answer to that question. So, and then I would love to know other things like you know, once we know why certain physical systems can be conscious or not, then you can ask and questions that I think that I would, I think that also are ethically relevant. So when, when are babies conscious, you know. When someone loses consciousness, right, and we make a lot of medical decisions these days based on whether or not we attribute life and consciousness to certain individuals or animals. You know, it would be important to, the reasons why we want to understand is because there is a physical, there is a mystery. But at the same time it has consequences in terms of the decisions that were made and ethical consequences, right. If a-- if a computer for-- or take a car. An auto-- you know, self-driving car is conscious and then kills a person, are we going to feel differently than if this self-driving car is not conscious. Are we going to make that distinction. Is it going to be the same liability, you know. Are you going to trial a-- you know a car that is unconscious. Or are you going to trial the engineer who actually you know, like, created the algorithm to create that. So it has, it has consequences. But the mystery is you know what, why, and how consciousness fits in the universe.
RS: The leading competing theories of consciousness today?
LM: I mean there’s many theories, right. And I I think when people say leading theories, I have a slight you know, reaction towards that, because that already attributes that, that some are better than others by some means and I don’t think that we can necessarily know that. We know there is a bunch of theories, we know that there is a bunch of theories that have more (INAUDIBLE) than others. So if you if you mean leading by how many people adhere to those theories, yes, there is leading theories. There’s people-- there’s theories that people take more at face value as being the right ones, right, than others. You know, so-- and there is many. So we have-- you know. There is theories that, and then this is part of the ones that I used to, or variants of those that I used to work in the past, that understand consciousness for instance as information exchange. You know. This is a globular workspace theory. And the-- the, the initial proponent was Benjamin Baars, and he was a psychologist and he was thinking more of the mind in a computational sense. So he was thinking of modules that we’re unconscious, that would process information, and then the modularity would be broken up by information exchange across those modules. And then Stanislas Dehaene and Jean-Pierre Changeux, and um-- I’m looking the name of-- (INAUDIBLE NAME). They extended that into mapping that into the brain. So then they assume that do certain neurons in certain areas that because they have long axons, they can connect and then they map it into the brain, right. That, that’s why it’s called neural loyal workspace theory, and not just global workspace theory. Then there is a integrated information theory or you know, it’s a theory that is being developed mostly by (INAUDIBLE NAME), you know in recent years or so by Christof Koch. And there’s a number of other people like Manuel Lee and Martila Mossmini, who I work on that. It’s a very different theory from the more computational theory that Stanislas Dehaene has-- Dehaene and proponents.
So they have a very different stance into how they define consciousness and also how they explain consciousness. And it’s the theory necessarily about the brain but it’s a theory about physical systems and what would it take for any physical system to host consciousness, right. And the attribute that to a system that have causal powers. Causal powers means that you know look I can take a difference and make a difference. And then there is high order theories that they think of conscious more as, I think I am in this conscious state. It’s more a philosophical branch. I think that the new biological underpinnings of that theory are more in the work. So it’s interesting because the theory that it’s been developed mostly by philosophers. And in the neuroscience there is only a few people working on that. One interesting variant of that is the radical plasticity hypothesis by (AXEL NAME) which he says is a variant of the high order series where he thinks that the brain learns to be conscious. What that, what does he mean by that is how you know he thinks and you have first order representations, say like color. But then over time the brain learns to map like, I think I am in this mental state. I think that I am. And he’s viewed this as, this is our learning process, right. Then there is others, which I’m not sure people would call series, but you know, predictive coding for instance, people are trying to think of how predictive coding which is a theory, it puts the brain the upside down, so creative coding makes the claim that in essence what the brain does is it tries to infer what are the causes of the sensory input, right. And the idea there is that you know you don’t have access to what the world is, you only have access to what you think the world is. So in their view what the brain does is just try to come up with hypotheses as to what could have caused this sensory experience. And there is proponents such as Andy Clark, trying to-- or Jacob Hallway to look at what would, and also Neil Seth, and trying to think of, how could predictive coding claim consciousness and especially, especially subjectivity and they are working a lot on their relationship between the viscera, thinking about the brain by the way, and consciousness. The viscera and, and the brain. So the idea is that your sense of subjectivity, and your sense of existence in certain ways, your basic sense of being could be related to how you feel your, how you send your body in, in light of your brain. And there’s some others, so there’s quantum theory. There is-- there’s a number of theories. So it’s not that we haven’t, not that we have few theories.
RS: The Templeton Grant? Why it’s important?
LM: So-- the-- there is many reasons why the Templeton Grant, why the Templeton program is important. One is, and this is just regardless of the field of consciousness, you know. That if you work in other domains, like I also happen to, to work on the issues of language, for instance, and memory. There is many in most of these fields, you have people holding different views. And the way in which we usually do science, is that we take the view that we like. We run experiments to kind of confirm the theory, and then we’re happy, right. But then there’s another guy who actually think differently and then he runs his experiments or her experiment, confirms his or her views, and they don’t talk to each other, right. So then at the end what happens is that we have like multiple, multiple threads of what I would call more than, more than theory, hypotheses, right, models, research programs if you want, right. And they didn’t, they don’t talk to each other and what is interesting is that very often they also make either contradictory predictions or sometimes there is data from one camp that contradicts the-- that contradicts separation from the other camp. But the other camp kind of tend to neglect them, right. And so I think that one of the, the very good aspects of this particular program is it is trying to make an effort to say ok, let’s just take two theories and agree to disagree. So what are the, try to map out the places where what I predict would be different from when you predict. So therefore we know-- therefore we can learn. This doesn’t necessarily mean that the theory that wins is the right one. And that’s an, that’s an important-- you know it’s important about subtle conclusion. It just says that the data are consistent with, right. And therefore it does two important things. So it tells you something about consciousness on the one hand. So this is probably, you know, if you say that for instance consciousness is in the thalamus. And then once the you replace it in the thalamus and your theory predicts that it does in the thalamus, right, and you find the thalamus and you say ok, well, the thalamus is important. And is consistent with this theory, so it would be worth continue investigating why is the thalamus and not something else, right. So it’s important in that, in that sense. I believe and this is not necessarily a part of you know what the, what the Templeton per se is putting emphasis. But I think that is what is enabling and then it cannot be underestimated, is the open science aspect that we are embarked on, right. And why this is important is because what happens currently is that labs conduct their experiment. They have this data and then there’s silo those data. Nobody else looks at those data, right. So then you cannot test other hypotheses on those data, right. And so then there’s a bunch of problems with that, so one is data are lost. And that is a waste of resources right. So if you think from there more like, larger perspective, and if you think of, you know, we want to use, have the most rational use of resources, you would want to recycle those data you have, right. For some other purpose. If you do open science, that’s what you do. So now you actually put the data out there, so then people who, people can actually look at these data without any theory regardless. And then look at ok, so what, so if this theory was right, then should also predict these other things. And you can continue with their scientific process you know for years to come, right. Well you can actually take it other theory regardless of the one, you know, we happen to be getting two theories. But somebody another camp, from a third theory, fourth theory, could take the same data and say, are these data compatible or not with my theory. Right. So we can, so is the, is there is the unions so to speak of the adversarial part on the one hand, but mostly in my view the open science aspect that makes them, that will make the most impact.
RS: Define open science for those who don’t know what it is?
LM: Open Science is, you know, first of all is making the protocols experiments data and analysis shareable. Open to the whole public. Which means that you know you put them in the cloud and everyone is allowed to download them, play with them, and we use them. Of course under certain-- you have, they have to sign an agreement, because of course you don’t want them to misuse the data, to try to trace the subjects back, and so on and so forth. But so long as they agreed or not misuse those data for a particular purpose, they can use it for any further discoveries. Now, that the open science comes with a lot of challenges, right. And that’s, that’s the most difficult aspect I would say of this project. Part of that is, you know, when I collect my own data, I know what I did. Right, so then I don’t have to specify to you like look, I collected this data under this circumstance, it was you know like, at this time of the day and the subjects had taken coffee and I asked them, you know, there’s a ton of information that as a researcher you know that you don’t put in writing in a paper, right. But, but you of course know that will affect the understanding of those data. Those are the so-called metadata, not the data itself but the metadata. And having data or sharing data that can be reused, it’s only as good as the metadata are. So if I don’t tell you in which context these data were collected, you will, I mean sure, I would share with you the data, but you won’t be able to use them to do anything, right. So the biggest challenge is how to you know, publish the data with a proper set of metadata that will allow people to reuse them. And these is a very difficult challenge. It’s, you know, I think yeah, it’s a difficult challenge and a lot, very often people don’t want it. Don’t want to deal with it. Like my students for instance, they, they feel like look, I already collected my data, I published my experiment. Now why do I have to invest like six months in just, you know, curating the metadata for others, you know, to profit. Because it’s not, you know, you’re doing, you’re literally doing charities so to speak. You’re not going to be profiting, it’s others who will be profiting, right. And in a very competitive environment which is science, this, this model of, you know, you go and undergo the whole, you know, you collect the data which are very expensive, you curate them, you give all of the information, then you know you have another lab that you know sits in their, you know, lab. They did nothing and then they can profit from that. So imagine how it goes with the other geoscience, right, then of course it doesn’t go really well. You know.
RS: For the consciousness project, you’ve chosen two competing theories of consciousness? How did you choose those two theories?
LM: So the process is, I did not choose those two theories, just to you know for direct to set the record straight. I was invited to a symposium where people from integrating information theory, (INAUDIBLE) workspace and a variant of higher order sort theory were invited. And the idea was to come up with experiments to test those theories. So the-- the realm of theories to be tested was already set up by the invitation process, and that’s something that I think that (NAME) and Christof chose from a very top-down manner. I don’t disagree with that. I think that you know they, they chose theories that have, that can be tested. And that’s why I think that, you know, it’s a good enterprise. So at the end, it turns out that we’re testing two theories. So one is Global Neuronal Workspace by Stanislas Dehaene, (INAUDIBLE NAMES), and Integrated Information Theory by Julian, Christof, and others. It was, in that, I belief that you have talked to Christof about this but in that process, the idea was that in that workshop to try and think about experiments that would arbitrate among these two theories, right. And arbitrating means which I think is important for people to realize is that, you make positive predictions. So theory A says I predict X. And theory B predict says I predict Y, and then you can measure both and you can say that look, under, under situations where we can measure both, it can know which one of those happened under these conditions or did not happen under these conditions, right. So that’s, that takes us a long time to figure out the protocols by which we would find, because you need to, what is important is to figure out those few cases where predictions make incompatible predictions because of course they make a lot of compatible predictions. But those are not, you know, they’re not informative for the process of, for the adversarial process. So one needs to map the incompatible predictions. So it took us quite some time to figure out protocols by which we could agreed on those incompatible predictions. Then took us also quite some time to figure out which techniques I mean as I mentioned before, the problem with the, with human neuroscience or we decided to start with, of course at that, at that meeting, it was still the question open, do we do these studies on animals or do we do these studies on humans. And we opted for starting with humans, under the following assumption which is that in humans we know or we can more or less infer that humans aren’t conscious, they will tell us, very often if you expose you and me to the same sensory stimulations we tend to give the same responses, so there is a way by which we can more or less assure that they are unconscious. With animals is much more tricky, right. They don’t, we don’t share the same brains, you know, that these things are wired differently, and even if they give us responses, it’s just much more, the inferential process much more difficult. So that’s where we start, we decided to start with humans. The price to pay of course is that the techniques that we have for human experimentations are not, in some ways are not as good as the ones for animal experimentation, right. So we decided, because the techniques are not perfect, to devise protocols by which we test the whole arsenal of, of techniques that we have in human neuroscience. So it would be FMRI, ENG and EEG, and also intracranial recording. So this tell you that this patient will have electrodes in the brain. And then the idea was that across all of these techniques, we could get a definite answer by knowing that okay, some techniques are better at giving certain answers than others. So in the decision process in terms of weighing like ok, this theory predicts X, this theory predicts Y, we would take much more-- we would take a specific technique to answer that question, right. So the theories, I didn’t choose them. (LAUGHS)
RS: Where are you in terms of the project now? One theory? Multiple theories?
LM: You know, there is multiple interests, you know, and I will start with the interests of the Templeton Foundation and then I will tell you like my own interests, you know. So the Templeton Foundation has the interest of trying to eliminate theories, so to speak. And by elimination what they mean is that you find evidence that, that is substantially compatible with certain theories, and incompatible with others. And by doing that you kind of like, live, you know, diminish the weight of those particular theories. So if you think of, you know, (INAUDIBLE) you know, idea of science, in that, in that research program, some programs will be more effective than others and the others will degenerate over time. So in this sense you would eliminate them over time. Which, you know, we will do and this is I think is a laudable aim. I believe that these, in combination with open science aspect, is what will give us even a fuller understanding. Why? Because on the one hand we will know because we would get a consistent evidence with one theory, that certain area or certain set of postulates are possibly, you know, true. But also, people would be to look at these data in multiple ways. So, because I don’t think that you know, put it differently, it could be that any one of these theories is right. But it could also be that all of these theories are wrong. Right. So, and because of that, I think that a combination between trying to eliminate those that are not particularly predict-- that don’t predict responses equally well, plus, discovery science, which in this case is being able to allow the whole, you know, research community, to mine those data for the time being, is what will yield a better answer. And one thing that you know we are, I am personally particularly invested in that aspect, the discover-- so putting together that theory driven aspect, which is the adversarial collaboration, with the theory free aspect. Right. Under having for the first time a data set which is curated on two experiments and on three techniques that does not exist currently. Not for consciousness, not for memory, not for language, not for any, not for any, you know like, function in the brain. So the-- the other best project comparable with, you know, that was small. But comparable is the Human Connection Project. The Human Connection Project only collected from FMRI images on individuals. There has been no project that collects on the very same paradigms ENG, FMRI, ECoG. Right. So this will allow us even to, you know, if you forget about consciousness for a minute or you or you put that on pause, it will allow us to even understand that how does electrical activity measured by ECoG, relates to my FMRI activity. And by knowing that fundamental question, I will be able to understand not just consciousness but consciousness, memory, language, and etc, right. So I think our product has a far bigger scope. You know, that we’re trying to understand you know what the brain does and also what, you know, what consciousness could be in the brain.
LM: It’s Electrocorticography. These are the electrodes that are implanted in patient’s brains.
RS: They are actually implanted?
LM: They are implanted. So they are literally on top of the brain. And that’s, you know, the only comparable recordings that we have of those are in animals. Right. So in animals, you know, this is, that it’s the best, it’s the best data that we could ever hope for, except for the fact, and there is always you know like this you know pros and cons that of course they come from people that have epilepsy, right. And then, you know, like then it’s, you know, you need to always triangulate and say ok, look, could it be that part of the epileptic brain is diseased but in the other parts are normal. And if the other parts are normal I can make conclusions from those normal parts. And if I then see the same response in normal brains and in brains with epilepsy and you say ok, look, these two things match, I can learn from it, right. So there is that, there is-- that we don’t have around. You know...
RS: The uniqueness of this adversarial collaboration could be used beyond just trying to understand consciousness?
LM: Yes, absolutely way beyond.
So -- put it differently. Because we are collecting data on two experiments and then three different techniques, and these will be open to everyone to mine for the years to come, they can investigate even like, you know, mysteries of the brain that do not necessarily pertain to consciousness. They could investigate things as, you know, what are the-- what are the underpinnings of the bold response for instance in FMRI. And that could, you know, trigger new discoveries, right. Well then you could ask, oh, ok, so what is the, what is the information that I can get through for instance, ECoG that is lost in EEG, that I can still recover on FMRI. So then you can ask the question, which techniques are better to tell me what we chose, still the questions we don’t have. So we will be able to understand consciousness better, because of course people can go and conduct further analysis than the ones that we have planned. Right. For understanding consciousness, but you can just take it for more like basic science. And the fact that this will be open to for everyone to do this, will multiply. So just think of revenues, you know. To me, this is the best revenue you could ever hope for, because now you the whole world be able to mine the data, you know.
RS: How unique is a project like this?
LM: To my knowledge there is nothing like this around. Period. Not just adversarial collaboration, that is an interesting part and that’s also unique in its own way, but having datasets that are replicate across two independent labs, on three different techniques that will be open, this is just unique. So the Human Connection Project, what they did for instance, is they took specific scanners that they tweaked and they, you know, systematized to collect those data. Right. So now we’re trying to face the next challenge which actually every researcher phase, which is how would you standardize across sites, right. And across different techniques and how do you put this together, right. So there is nothing like this around. Not in the human (INAUDIBLE), not in-- you know, simply there’s nothing. And that’s exactly why I’m so invested in doing this. Because I have a feel that what Templeton is mediating is, or is giving us the key to do new things. And not just for consciousness, for conscience and beyond. But beyond in a way that you know we don’t have that currently. And I’m also, you know, it’s, it’s very challenging to do what we’re doing, you know, because you know, doing (INAUDIBLE) across labs, doing as he told you like for instance, collecting the metadata so that people can use these data later on, is also extremely challenging. And you know, a lot of the technologies also are not developed for that, so we are, we are we’re having to develop electronic lab notebooks, to even do that. You know. So this has required a lot of development that goes far beyond this specific experiment that we’re doing, right. But I do think that once we have that down and we show people how this can be done, my biggest hope is then we will show people and researchers a process by which they can continue doing this. You see. So what I, what I would really hope is that this is the first project and then you show like, ok, if you want to do this, you have to do all these set of you know like, standardizations, you know like here’s the electronic lab notebooks, this is how you create the data. Now go and do it for another consortium. And by doing that we were going to get more and more data, right. And I’ve discussed this with (INAUDIBLE NAME) because what my hope is what I call Il Alma Laboratory. So Il Alma, it’s a, it’s a metaphor of the large scale telescope that is in Chile. So what do they do, they have these, you know, much smaller telescopes that what they do is that they put them all into one, like they align them to look further, so to speak. So now imagine with the same with the brain, right. Now you collect data in your laboratory, I collect data in my laboratory, or it could even be like, crowdsource experiments. I said look, I have this great idea, who signs up. And then Stanford’s lab, Harvard’s lab sign up. We all collect data, we all share the data, and then you know that we, we tackle a problem. We don’t have the technology know to do that. And what Templeton is allowing us is to develop those that process and the technology to do that. So it’s going to be far beyond this, you know, adversarial collaboration.
RS: You raise an intriguing point about whether babies are born conscious or learned consciousness. Where do you come down on that debate?
LM: The initial question is, are babies conscious or not, period, right. So that’s, that’s the first mysterious question and we don’t know the answer to that. And again, part of that is like, if you think of behavior. Babies don’t have the same like behavior, you cannot instruct them necessarily the same way you would instruct and adult. Which is more challenging. So there’s that. Second is, there is certain people who think that consciousness is a process that is to be learned. So this is the (NAME) in particular, so he thinks that there is a process by which the, in particular like, high order areas, in this case you can think of, you know, prefrontal cortex and so on and so forth, learned to look at first order representations. And we know that for instance in babies, that more like sensory areas are mature at birth, whereas the frontal cortex actually start you know, it takes much longer for the brain to mature in those areas. So there is also changes in brain, in brain connectivity, right. So, you know, it’s, it’s thinkable that, that it changes over time, right. One possibility is that it changes in its form. So for instance, and this is (INAUDIBLE)-- my former you know PhD adviser did. She was looking at what kind of language capacities babies are born with, right. Because of course you know that when, you know, you grew up in the United States and you speak English and you are sensitive to certain phonemes. I grew up in Chile, I speak Spanish, you know. And I am sensitive to other phonemes. Somebody who, you know, who grew up in Japan is sensitive to their phonemes. It turns out that babies have the capacity to react to all of the phonemes. But as they’re exposed to their mother tongue, they lose those capacities so to speak. Right. So in Spanish for instance, B and B. These two con-- these two vowels, these two consonants, they get mapped into the same sounds, so to speak for us. So we cannot even, I cannot hear the difference. Or in, you know, for-- for Asian speakers or for people that speak a language, the R and the L for instance, they are not different, right. So you lose that capacity. So in essence, we do know that certain aspects of our conscious experience are lost. As, you know, we, we you know. So it’s not the same thing, right. And you can also think that there’s, or some words we grow, you know, to have, to think of you know, maybe at the very beginning babies do not see faces and they probably have to learn to see faces. Maybe what they see are just shadows. But as they get exposed they learn that pattern and then eventually you just see face. You cannot, you know, the (INAUDIBLE) face experiment. So whereby if I show you a gray and black patterns, sometimes people think that they, as if they are blobs. But then they show you the face, real image, and then they go back to those blobs, and then you’ll see a face there. And of course you know the colors are the same, but now you have this category that you know somehow your brain learn and then poses it. And now you see a face. Right, so there, of course there is a part of the content that changes as we grow older. And babies for sure learn that. Now, did they learn to have the conscious capacity as such, like have the experience, remember because we were talking about of course the content, you know, that can change, right. But just having the experience are not, possibly they just have it when they’re born. For all that we know like their brains are pretty similar to ours. Right. Except that the connection with the prefrontal cortex, right, and that is you know, strengthened over, over the years.
RS: First order versus higher order?
LM: So they defer because they-- you know, the first order series think that having the content as such is what mediates an experience, right. So if you have for instance, you know, before and there’s activity in that area and you will see color. What people in the higher order are in I think is that, no. Just having a response in the brain is not enough. You need to have a receiver so to speak, like an eye, like you know like a mind’s eye that says okay, looks at that, that representation and only if it looks at that particular representation then consciousness is, is possible. Now why is that debate relevant, is because it gets mapped into different areas. So some people attribute responses in sensory areas as the predominant, as having a predominant relationship to consciousness. Whereas people in the high order thinks that you know, it rests more in the prefrontal cortex, how a predominant relationship to consciousness. Right. Which one is right or wrong, we don’t know. That's part of wanting to figure out, right? Does the prefrontal cortex have anything to do with consciousness. Or with only certain parts of consciousness, so imagine that for instance, it gets only there when you can, when you can think that you have an experience, when you can, when you can have a mental representation, like-- in a way like, you know that you know, but you know for a lot of the experiences you may say like no, that doesn’t necessarily (INAUDIBLE) there, right. So if you are in a car accident and you are very stressed, your experience is very vivid, but I’m pretty sure-- am I having this experience. Right. Or if you’re having an extreme pleasure or you know some people in meditation think that you can completely lose the sense of agency and the eye. I don’t know I have never done meditation or under those states but some people claim that you know that can be dissolved. Right. And all that is left is just experience. So it’s important for us to understand what, you know, what it is in the brain.
RS: Why is the prefrontal prefrontal cortex very important to understanding consciousness?
LM: I mean it’s more than it is important to understand conscious, it’s that certain theories attribute you know, attribute to that area of particular functions, especially in relation to consciousness, right. So it’s not that I wouldn’t say that V1 is more important or you know like prefrontal cortex more important, we know that the whole brain is important in certain ways. And we know what the prefrontal cortex is very important for many things, right. So for reasoning is very important. You know, if you want to inhibit a, you know, your inner drives for instance, you know like think of the, the conscious, the super ego, like you know it’s more prefrontal. Reasoning maybe more prefrontal, you know, anything that requires more, you know, to be more focused is more difficult, may require prefrontal cortex in some ways, or you know, things in working memory. So if you have to remember, you know like a telephone number, then neurons in prefrontal cortex are very important. Now do they relate to consciousness, we don’t know, right. There is evidence that points to the fact that it does, right. But there is other evidence points to the fact that it may not or, or say, or said differently. Many years ago with a PhD student of mine would realize is, in the-- in the 1990’s, at the end of 1990’s, Christof Koch and Francis Crick, they put the idea of let’s just concentrate on the neural correlates of consciousness. And the method to find that was very simple, is this whole-- abstraction method. So you have, you consider a situation where you are conscious, versus one when you aren’t, right. Then you subtract and you said okay, whatever is left in the brain from that contrast has to relate to consciousness. Right. It seems very simple. So in one case I am, take off like sleep. I am awake you know and I have a conscious experience, I am asleep and I’m in deep REM and I have no conscious experience. If I contrast these two, voila, whatever is there in one state versus the other would relate to the neural correlates consciousness. The problem is, it is not that simple. Why is not that simple is because there is activity that precedes a conscious experience. And that may also change. But it’s nothing to do with consciousness. So for instance, if I have an expectation, that might mediate the way in which we become aware, right. But of course you know like, it could get to the same state by other means. Like because I pay more attention, so there is some, there is some mediators that facilitate having an experience, but in and of themselves do not related experience as such. And in some conditions you might, you know, when you subtract and you might get them. The same thing happens with the so-called consequences. Right. And the easiest consequence is for instance, episodic memories. So think of the following case. So there was a famous case called HM. This was a patient that you know, he’s suffering intractable epilepsy, the hippocampus and you know areas around were resected, that patient never ever in his entire life was able to form new memories. You know, he-- he lived at, in the time at which you know, like he had had, he had had the surgery. And he could never remember like any new fact after that moment, right. Now what we know is that, you know, HM had experiences, he says he couldn’t remember, right. But under normal scenario, when you actually do have your hippocampus, we also remember things that we have been conscious of, right. And there is someone to one relationship in many, in many ways between them what episodic memories, like now, this moment, they if you are not aware of this, will not form a memory of this. But if I am aware, I might form a memory, right. So these are the so-called consequences. And they also will come because if you think that the conscious mediates some other process down the line, in these you know simple abstraction they will also come by, right. So that when it comes to the prefrontal cortex, a lot of people are saying that a lot of the, or some of the results that we are seeing, reflect the consequences of conscious as opposed to consciousness as such. And that’s a huge debate. You know. So it’s not that the evidence doesn’t show the consciousness or some experimental problems showed activity in the prefrontal cortex, they do. It’s just that some people interpret that activity as being a consequence.
RS: Does consciousness help us understand disease of pathology or perhaps the other way around, diseases pathology helps us understand consciousness?
LM: Well I would say both, both ways, right. So I think of epilepsy. Why, you know, what happens very often epilepsy is that people lose consciousness transitorily. Right. And we have, we have tried you know, one of the reasons why epilepsy is I think is an interesting model is because we have a transitory loss of consciousness and then we ask, what happened in the brain at that time, that made this subject to lose consciousness, right. So in many ways it will help us, by understanding the diseases, we can understand what, you know, whatever responses are that mediate consciousness. It’s the same thing if you think of you know, brain injury. If you have injuries in certain areas of the brain, you sometimes will have deficits like HM, you know, that’s a very good example. So we know that you know he had memory, he had memory deficits. But he did not have consciousness deficits. So now we know that the hippocampus is most likely not necessarily really relevant for consciousness, because you can take it out, of course you lose memory, but you don’t lose your experience. So in that sense it helps us, right. Will consciousness help us understand disease. If you think of you know, for instance, schizophrenia, right. One of the problems that the people when, when they suffer from schizophrenia have is like hallucinations, right. And for them, those hallucinations are real. You know, it’s not that, you know, when they say look I am seeing X, it’s not that they are bad people, they are really seeing X. Right, and they suffer from that. And we need to understand, you know, and help them. How can we remove that. Or for instance, people who suffered tinnitus. That is very common. People whose suffer tinnitus is like a ringing in their ears. And grow, it can grow really, really, really, really loud, you know. And people suffer from it significantly. So if we want to understand like what causes that, we could help them, right. There’s other people on their way, you know, that for instance, they’ve lost consciousness because an area of their brain was, you know had an, an (INAUDIBLE). Those people suffer because they can, this is the so-called blind side, you may have heard about this. So these are individuals that if you ask them, can you tell me whether these bars that I’m showing you in front have a stripes due you know to the left-- have stripes horizontally or vertically. And they say, but I cannot see. And you tell them, ok, fine. Never mind. Can you just guess. Sure, I can get guess. And then they are very good at guessing. So they actually can tell you very well, you know, whether the bars were, you know, horizontally oriented or vertically oriented. Now they suffer because they cannot see. What does, what does that mean, it means that they don’t have a conscious experience, even though the information is there, their brain processed it, they can even you know look at other behaviors, they don’t have the experience. They suffer. So I think that all of these things are intertwined, right. You know, and also if you think about it, even for, we discuss this for medical decisions, right. So if you are not conscious, for instance, if you are, if you’re under (INAUDIBLE), and you are completely unaware, and you commit a crime, what happened then. Do we try and do the same way or not, right. If you commit a crime because you have, you know, hallucinations. We don’t send you to jail, right. We send you to some mental institution, right. So there are many ways in which all of these things are intertwined.
RS: Beyond human, other species and consciousness?
LM: That’s an excellent question. So this goes back to you know. That’s why we would want to have a theory, because we don’t, we don’t, we we cannot be sure and we discussed this before with, I can only be sure of my own experience, I cannot be sure of yours. I can only infer it, and you know, it’s a good explanation because you seem to move the same way, you seem to be constituted the same way. But with other animals it’s a little bit more tricky, right. So we rely on behavior. They don’t, they don’t talk so they can’t really tell us, we just have lasting knowledge. No they can’t explain or described experiences. So we rely on some behavior, we have to look at, a human would behave under this circumstances this way, right. So for instance, I told you about the blind side, that’s how very, that’s a very good example. So (INAUDIBLE NAME), many years ago, came up with the following exam-- with the following paradigm to see whether or not a monkey would be conscious or not, or would be blind side or not. So then she presented in the Emmy field where the monkey was, was conscious, because these deficits are usually in quadrants or, or in any field. So it’s not the whole visual thing so just certain parts, but she presented in the sighted field a stimulus and asked a monkey to say ok, is this left oriented or right oriented. And sometimes she would present the stimulus and sometimes she wouldn’t present the stimulus, right. And then she would ask the monkey, in all of the situations, is it left or is it right, is it there, is it not. And sometimes she wouldn’t present anything. But even though she would ask the monkey to say is it left or is it right. Then of course when she didn’t present anything, the monkey would be a chance, because it was nothing there, right. And in the sighted field, the monkey would say, I didn’t see anything, right. And in the cases where he, when she did present something then the monkey would say, I did see then the behavior you know would be consistent with that. Then she did the same thing in the unsighted field. In the part where the monkey presumably didn’t see, right. And then again the monkey you know would be able to say is it left or is it right, but then the monkey would always say, I don’t see it. So that by doing that, then she said okay, you know, we can infer here that as that monkey behave similarly, as the humans would behave that the monkey is not seeing. Is she sure about that. No. It’s a good explanation is reasonable, right. It’s a big difference, right. But without the theory in a way that would explain us why it, consciousness, should have been there. You know, we wouldn’t, we would never be sure. So, how far are we understanding how, which other animals are conscious, I mean not that, we’re pretty far because we don’t have a theory that explains that necessarily, right?
RS: On Earth, are humans the only conscious animal?
LM: So I don’t know the answer to that question but I have, you know, beliefs. And my belief is, if I think that conscience is in some biological matter, like you know what I have, then I think that it ought to be there for other species, right. You know, it’s just simple, there’s nothing that we think there is like a, you know, evolutionary jump, right. So it’s, it’s very reasonable to assume that, you know, it’s there, you know, like in dogs, is there in monkeys. As I said if you ask monkeys or other animals to be, to, if you put them these tasks that you know humans would behave in a particular way, they tend to behave the same way. Right. And not only that they tend to behave in the same way if your lesion certain areas. Right, so if you would, if you were to lesion for instance the hippocampus in the monkeys, then they wouldn’t have these behaviors. They would only have the behavior if you lesion an area AV1, which is the same area that you lesion in humans. So kind of this homologous. It seems that across species we have homologous, right. And so, so long those homologues exist, I believe that, you know, I have no reason to believe that I am superior, so to speak. Why should I, I’m not something else, right. You said the question is more like how, how far down the animal kingdom you go.
RS: Technology, one day machines will be conscious?
LM: I would probably answer the question the other way around. We need to know what consciousness is to be able to answer that question, right. Because in principle I don’t see why not. So if at some point we understand what physical systems afford the capacity to have consciousness. And we can replicate that the very same physical system, I would say ok, there’s homology, right. So if there is the same thing why not, right. It wouldn’t, my inference would be it should be there, right. If I don’t have that, then what is going to be my conscious meter, because I do know that you know computers are built of you know, chips and so on and so forth, they don’t have this physical matter, right. They are not (INAUDIBLE). Right. So-- would they, they might be able to do the same thing. And that’s the tricky part, right. But this doesn’t entail that they are conscious, right, as he said behavior, so to speak. You know, or having certain, you know, being able to categorize things doesn’t mean that you see. If you think of deep neuro level these days, they can tell you really well was it this face or that face. Do they have the conscious experience of a face? I’m not sure about that. They can just run a-- super good categorization you know machine, that’s it.
RS: So you don’t worry about for example, Hal?
LM: Well so, yes. I do, I do think that a bit differently. I, for now, because we know that biological feature-- biology does seem to support consciousness, I think it is reasonable to assume that. Is it the only thing that can support consciousness? Maybe not. So what I’m saying is that so long we don’t know what is, what are the characteristics of any physical matter that are for, that allow for consciousness, we will not answer that question, right. Now on the next aspect of your question, Am I, am I worried about machines overtaking us. Well in some ways I am but not because they might be conscious, you know, but imagine that you know they become much smarter than you or me and they can predict the, you know, stock markets. So they start willing, because of other things you know, so I don’t think that is necessarily because of their conscious, it’s because of the things that they can do, right. You know so imagine they control the stock markets. You know, they will be able to make your life and mine you know very miserable, you know. Or-- Yeah. So it’s because of-- or you know, self-driving cars, you know. Imagine that eventually they decided you know they want to kill everyone, because there’s a bug in their program. Of course I worry about those things, they don’t even moral. So if you don’t have conscious you might even have moral, so do you suffer if you kill me, I’m not sure. I mean, you do. I mean hopefully, you know, but I’m not sure what a car would suffer. You know.
RS: Autonomous vehicles, autonomous machines that are making decisions on the spot that could have moral and ethical implications without a sense of consciousness. Where does this take society?
LM: Well that’s, I think that’s something why we need to, you know, there’s many reasons why understanding consciousness is important. One of those is for the medical decisions that we make. But the second one is for the ethical decisions that we also make. Right. So are we, you know, who we hold responsibile as I said to you before like, is the engineer who programmed the car. Is the car itself. Do you, you know, what do you do if you know, I mean it happened before, right. Tesla actually did run over a person, right. So, do we trial Elon Musk, you know? After all he’s going to keep putting this thing out. Right. Or there is you know, all of these Boeing’s. You know, are completely landed. Why? Because you know the autopilot you know, it was not necessarily being well tweaked and it led to a lot of, you know, loss of lives. Who do we trial there? Presumably not the, you know, airplane. The individuals that did not take enough precautions on testing you know like their machines. Right.
RS: They’re not imbuing consciousness into these machines?
LM: Correct. And I think that's probably part of the problem, right. So I think that eventually, you know, maybe what you want is to not only create smart machines that can do you know like translation really well. Right. But you want to imbue-- put consciousness into those machines such that they would feel moral responsibilities. You know. They, you know, under certain cons-- under certain circumstances they would say, look, I actually opt not to act right, because otherwise, I mean unless you can have such as smart algorithms that would also put all of the ethical dilemmas and you would program it. That's also possible. And then you wouldn’t, then you can dispose you know from consciousness, right. But the problem is that with ethical dilemmas they are difficult, right. So they, we don't know. Do I kill one person if I am driving the thing. But then I killed three if I don't prevent something, you know, these are the moral dilemmas and you know we humans have to deal with them. Would you program that, that solution. Is it even ethically responsible to program that solution.
RS: As you peek into the future, is this resolvable? Or many generations worth of work? The big questions?
LM: So it’s-- I think you’re making-- your question is very good in the sense that I would love to have an answer, in the sense that, I don’t think, I cannot foretell what’s going to happen, in principle, right. I would hope that we can make substantial advancements. And that’s part of the reasons why I embark in this area of collaboration and in doing these open science is because I think that if we change certain practices, we can accelerate progress. So in that sense I hope that we will be able to accrue more information and get much more satisfactory answers, while I still am, while I’m still doing research. If we continue doing what we have-- if there is no advancement on technology, will we solve it. I would be less hopeful. I do think that we need technologies, right? We do need technologies by which we can not only measure neurons, and a bunch of neurons, but technologies by which we can measure how does one neuron talk to another neuron in another area. Right. Because now what happens is that we only have access to one area and we can record a bunch of neurons, then another area we can record a bunch of neurons, but we know that, imagine that you have, I have a language here, then I do some process by, by which I transfer the information there and then gets here in the receiver, maybe actually everything happens in the connectivity, in the connections between these two. So you need to really map who is talking to whom. Right. And by which means. And that is something that we don’t have yet. Right. So we need to do that. We need to hopefully be able to manipulate selectively in the sender and the receiver. And in many areas, right. So you could actually say like well, if I knock out you know three of these neurons and hopefully can map which neurons those are, because the next problem which is that, not all of the neurons have the same neurons, I mean they talked to you before about these layers of the cortex. So when we talk about the cortex we think they say one thing, no. The cortex has different layers and there’s been, these many different neurons in these layers, and where the information goes from those layers, is also different, right. So then maybe if we, you know, selectively deactivate the neurons, the optical neurons, then or the optical dendrites, will have a different response if we do it at the bottom, right. And these are all of, you know, we need technologies by which we can look at the layers...
By which we can selectively inactivate them, right. And also across areas, right. So there’s a lot that will need to be done, right. The good things that actually people are, as I said, there’s a lot of initiatives these days, in creating new technologies, especially for this kind of questions. so I do think we’re going to, we might be, we might have tremendous breakthroughs, right. And if you think of crisper and you know, the genetics, they could revolutionize what we can do now. We have now this (INAUDIBLE), now we can grow, you know, like a set of neurons and we can connect them in a particular way, and then not only just have brain, you know like, you can grow things in a particular way and ask, now if I record, what happens. So, there is a, the explosion of technologies is, is amazing these days, and it-- if we pair with that, the answers might be not, a far reach.
RS: If we crack the code of consciousness, how does that help us understand us as humans?
LM: This is well, I will actually put that way around, if we are not conscious, do we care. So I would say that understanding consciousness is at the core of our existence. When I am not conscious, I don’t even think that I exist. So, like, you know, if you ask me like yes, this is what we need to understand, it’s the basic thing. It’s what makes all of this to be there. Without consciousness, nothing of this exists. Right, so in a way it’s just, it’s key, right. Now, but there’s a different part of that question, which is would it help us to understand humans, you mentioned. And they are not sure, right. I a way you could say, I don’t think the conscious necessarily is uniquely human. Maybe language is uniquely human. Right. Consciousness might not be. Yet, if I don’t have consciousness, I don’t have language and I don’t have a lot of other things, and I don’t suffer and I don’t care, I wouldn’t even ask the question whether I am conscious or not. You know, nothing would exist.
RS: Why do we need to understand consciousness?
LM: Well, you know as I said, the-- if you think of consciousness as what gives us existence as such, then it is what we are, right. So we need to understand that to understand who we are. Period. It’s very basic. So in that sense, that’s why it’s the biggest mystery, because it’s what makes us exist. Or we would be zombies, right, or probably computers. You know, that could do all sorts of stuff but wouldn’t know that things are there. Like, I’ll give you-- my best example is the Chinese room of Searle. This is from the domain of language. They were talking about what it means to understand language, semantics, which in many ways related to consciousness. But then he said well, this is a Chinese room, so imagine that there is somebody who doesn’t know Chinese whatsoever. But sitting in one room, it understands German, for instance. And then it has like a mapping set of rules, by which if you have this simple, you translate, you pair it with this. This is simple, you pair it with that. So then you give me a set of sentences in German, I pull up this pairing and I give you, in a, through a door, the answer. So seemingly I’m translating and seemingly I am understanding Chinese, right, to you. But of course I’m not. So like, that doesn’t exist for me. The only thing that exists is German. The only thing that exists, I’m having certain behaviors. That’s consciousness, you see. Things that we don’t experience just don’t exist.
RS: Cracking this code would help solve one of the biggest mysteries of life?
LM: For sure. Yes, yes.
Yeah, I mean this is the biggest, understanding how consciousness, which is like, in some ways is non-material, how does this property get into the material world. Is the biggest mystery. And some people say it’s such a mystery we will never like, it’s out of our reach. Some people say look, don’t even spend time on this question because it’s so difficult, that you know, we would never understand it. Right. The system is (INAUDIBLE), really serious problem. It’s a hard one, yes. But it’s really the most important thing we probably want to tackle. For sure.
RS: Did we miss anything?
LM: I don’t know-- no. One thing-- I’m not sure it’s important is... the replication crisis for instance and how we are also tapping into that.
The so-called replication crisis. So there is it’s, what has happened, not just in psychology or neuroscience, but in many fields, is that people put out results, findings, and then someone tries to replicate then they can’t.