Theories of Everything with Curt Jaimungal - Aaron Schurger: "No. Neuroscience Does NOT Threaten Free Will."

Episode Date: October 30, 2025

The Libet experiment didn't kill free will, says professor of neuroscience Aaron Schurger; the brain's "readiness potential" is simply stochastic neural noise. We then tackle consciousness's role in initiating action, the hard problem, and what it'd take for neuroscience to truly disprove our choices. If you’re interested in the topics above, I think you’ll love this podcast. Sponsors: - Get 50% off Claude Pro, including access to Claude Code, at https://claude.ai/theoriesofeverything - Get your copy of Richard Fain’s Delivering the Wow—available today on Amazon, or wherever you buy your books. - As a listener of TOE you can get a special 20% off discount to The Economist and all it has to offer! Visit https://www.economist.com/toe Join My New Substack (Personal Writings): https://curtjaimungal.substack.com Listen on Spotify: https://open.spotify.com/show/4gL14b92xAErofYQA7bU4e Timestamps: - 00:00 - The Readiness Potential - 07:14 - Reinterpreting Libet's Results - 24:18 - Autocorrelated Neural Noise - 36:06 - Noise vs. Conscious Imperative - 43:46 - Resolving Libet's Paradox - 50:55 - The Point of No Return - 56:11 - Attention Schema Theory - 01:03:47 - Explaining "What It's Like" - 01:17:00 - Consciousness & Action Initiation - 01:24:54 - Testing Consciousness Thresholds Links mentioned: - Aaron Schurger’s Papers: https://scholar.google.com/citations?user=lBSjfagAAAAJ&hl=en - Daniel Dennett [TOE]: https://youtu.be/bH553zzjQlI - Determined: A Science Of Life Without Free Will [Book]: https://www.amazon.com/Determined-Science-Life-without-Free/dp/B0BVNSX4CQ/ref=sr_1_1 - Robert Sapolsky [TOE]: https://youtu.be/z0IqA1hYKY8 - An Accumulator Model For Spontaneous Neuralactivity Prior To Self-Initiated Movement [Paper]: https://www.pnas.org/doi/epdf/10.1073/pnas.1210467109 - Consciousness Iceberg [TOE]: https://youtu.be/65yjqIDghEk - Stephen Wolfram [TOE]: https://youtu.be/0YRlQQw0d-4 - Awareness As A Perceptual Model Of Attention [Paper]: https://grazianolab.princeton.edu/sites/g/files/toruqf3411/files/graziano/files/cog_neurosci_2011b.pdf - Cortical Activity Is More Stable When Sensory Stimuliare Consciously Perceived [Paper]: https://www.pnas.org/doi/epdf/10.1073/pnas.1418730112 - Propofol Anesthesia Destabilizes Neural Dynamics Across Cortex [Paper]: https://www.cell.com/action/showPdf?pii=S0896-6273%2824%2900446-X - Jenann Ismael [TOE]: https://youtu.be/7kvXihDAOi0 SUPPORT: - Become a YouTube Member (Early Access Videos): https://www.youtube.com/channel/UCdWIQh9DGG6uhJk8eyIFl1w/join - Support me on Patreon: https://patreon.com/curtjaimungal - Support me on Crypto: https://commerce.coinbase.com/checkout/de803625-87d3-4300-ab6d-85d4258834a9 - Support me on PayPal: https://www.paypal.com/donate?hosted_button_id=XUBHNMFXUX5S4 SOCIALS: - Twitter: https://twitter.com/TOEwithCurt - Discord Invite: https://discord.com/invite/kBcnfNVwqs Guests do not pay to appear. Theories of Everything receives revenue solely from viewer donations, platform ads, and clearly labelled sponsors; no guest or associated entity has ever given compensation, directly or through intermediaries. #science Learn more about your ad choices. Visit megaphone.fm/adchoices

What this explanation offers is an answer. Suddenly it just hit me like a ton of bricks, and I was like, whoa, this is different. For decades, neuroscientists have claimed your brain decides before you do. The livid experiment seemed to kill free will. You're a puppet of your unconscious. Done. Simple. Right? Well, professor of neuroscience, Aaron Scherger, on the other hand, was not convinced. He thinks something fundamental about the most famous experiment in neuroscience is wrong.

I'm Kurtzai Mungle, and on this channel I interview researchers regarding their theories of reality with rigor and technical depth. Sugar says the readiness potential reflects stochastic drift towards a threshold and not commitment. Even more radically, the professor is testing his framework by examining whether consciousness has a higher, steeper threshold. we explore what would genuinely threaten free will, why the professor believes free will has not been killed, despite all of the headlines to the contrary, and we also explore new avenues neuroscience has to explain consciousness. Today's episode is a huge treat,

and I'm super excited for you to hear it. Professor, it would be great for you to outline the prevailing view on readiness potentials and its relationship to free will among your neuroscientific colleagues, and then where you diverge. Great. So for the viewers who may not know what the readiness potential is, it's this slow buildup of neural activity

that you see over motor errors before a spontaneous self-initiated action. And slow on neural time scales, it can stretch back in time up to a whole second or more before the movement, before the movement is initiated. So the prevailing view has been that that buildup of neural activity beginning quite early on before the movement is a reflects a decision that happened in the brain,

reflects a commitment to initiate a movement, and the brain is now getting ready to execute that movement. That's been the prevailing view of the readiness potential. And the reason it's been controversial, because of Libet, who asked his subjects to report when they felt the urge to move while the readiness potential was being recorded,

and showed that it looks as if the urge to move manifests very close to the time of movement, maybe 200 milliseconds before, 2 tenths of a second before the movement, whereas the readiness potential, which is taken to be the sign of the brain, actually deciding to get the movement going, started well before that. So that's the prevailing view, right? is that from a conscious point of view, your conscious decision or your conscious urge to move is sort of an afterthought

with regard to your brain actually preparing to initiate that movement. Okay. And what do you think? So I thought a lot about that result early in my career when I was a PhD student and even afterward. And I wasn't really studying movement initiation at the time I was studying perception

and working on perception. attention and consciousness research. But that result always struck me at a gut level. I just always felt that there was something seemed off about it. But I couldn't pin it down. And then one day, it sort of dawned on me, I don't know how or why, that, well, wait a minute, we're all taught, even in high school, that correlation doesn't imply causation. And this is clearly a correlational relationship between the readiness potential and movement. So what if it's not the cause of the movement? What if, right? Is it? Did anybody ever bother to ask? And that's when I started to think in more depth about it and like, well, how could it, if it weren't the cause of movement,

why would it be there? How would it be there? And then that's when this idea started to develop that maybe the readiness potential reflects the leading up to the decision rather than the outcome of the decision, right? Activity in the brain that's leading up to the decision. And in a subtler and way, in a deeper way, a sense that, well, what I did was I tried to come up with a thought experiment or an analogy that would explain that. And after a few attempts, I came up with a couple and the second one, I think, works best. There's the idea that if I asked a bunch of people to wear a health monitor on their wrist

during flu season, and I was interested in what happens, you know, when you get the flu, and in particular what happens leading up to when you get the flu. So I asked all these people, you know, a cohort of volunteers subjects to just wear this health monitor on their wrist, and if they come down with the flu, they should just press a button on the health monitor that records, okay, I have the flu now. And then I would collect those at the end of the

flu season and analyze those data, right? And one thing that you could very well find in that context is that if you align the data to when people press the button, you might well see that their health was getting worse before they even came in contact with the virus. Before they even come in contact with the virus. And of course, that sounds absurd. It's as if their immune system knew they were going to come in contact with the virus and said, well, why don't I just start getting sick in advance? Yes.

Right? The answer to that absurdity. apparent absurdity, is that no, because you don't get the flu just because you came in contact with the virus. You get the flu because you came in contact with the virus and weren't effective in fighting it off. So there's a tendency for your health overall to already be a little bit on a downturn before you even come in contact with the virus. If you look backwards in time from the moment when you declare, oh, I now have the flu. Right. So something that seems absurd, if you look at it from a different way, it's kind of a selectionist explanation in the sense that coming down with the flu necessarily selects the preconditions that are necessary for it.

And so if you look back in time, you're guaranteed to see something building up before it even, right, before by any logical sense it should. Okay, so let me see if I'm understanding this correctly. you're not disagreeing with the results of the libid experiments you're disagreeing with the interpretation absolutely yeah the results are very replicable you know i've replicated it many times lots of people have replicated that basic result the readiness potential is what it is it stretches back in time quite quite early but the the idea that i'm promoting is that it doesn't reflect the outcome of a decision but rather it reflects the the buildup the lead-up to a decision to initiate a movement and that that actual decision, that commitment

to initiating the movement happens in the context of a spontaneous voluntary movement. It happens, you know, maybe 150 milliseconds before the movement actually begins, so quite late in the game. Okay, so again, let me just see if I can summarize this for myself, but also for the audience. if you catch a ball, so someone throws you a ball, you can catch it, and that's reactive. In that case, you're responding to an external stimulus, and there is no, quote-unquote, readiness potential. If instead you spontaneously decide to throw the ball yourself, then that's what's called endogenously initiated. So it's generated from within. It's not triggered externally.

It's not triggered by an immediate external stimulus, yeah. Now, scientists like yourself and Libet use EEGs, so something on the outside of your brain to measure activity, electrical activity, and then they can record these spontaneous movements, and then when they do, they detect a sort of a preparatory signal, I believe it's called, if I'm not mistaken, and builds up from the top of your brain and the motor, the supplementary, if I was my research. motor area, yeah, yeah, areas that are kind of like, yeah, pre-motor areas, yeah. So this preparatory signal is called the readiness potential, and remarkably it occurs approximately one second or so prior to your actual movement or prior to you being conscious that you're deciding to move, and this is then taken as evidence of if something is occurring unconsciously prior, then it's not consciously prior, therefore that decision was unconscious, so you, unless you are your unconscious somehow,

then you did not create that decision. It's somehow a post hoc confabulation. So that's the ordinary... That's the ordinary prevailing view. That's correct. Okay. That is correct. And I think what is mistaken,

there or what was missed in that view, is that you can have things that reliably precede an event that aren't the cause of the event, that aren't directed at creating that event. There are various examples you can think of, and one of the, but I want to stress, one of the main problems

and the reason that you come to that, that you would come to that conclusion are to be found in the way that studies of spontaneous movement are conducted. And that is that you allow people to make movements spontaneously, you don't give them a cue to move, right? Like in Libet's experiment, he said, just, you know, wherever you feel like it, just press the button or flick your wrist, whatever the movement happens to be.

Now, what do you do? Right? You have this data. So how do you look at the data? Well, you find the moments in time when a movement happened, and you look at your brain data before that, right? So you've now selected out a sample of data where, every sample of data you select out ends with there being a movement. So you never looked at what

happens when there isn't a movement. And that's the problem. Okay. So just like with the example of the flu, when I collect all my health monitors at the end of flu season, right? What do I do with all the subjects who never pressed the button, who never came down with the flu? Toss that data. I don't analyze it. I just look at what happens. I keep the ones that actually did press the button and that's all I look at, right? That's where you can convince yourself that there was an early decision and an absurdly early decision when in fact there wasn't. And I think a good analogy is the relationship, for example, between barometric pressure and the onset of rainfall. So yes, if you look at every moment that rain starts falling in this location, you will, before

at hand, see that barometric pressure is going down. But what you've missed is that lots of times barometric pressure fluctuates either up or down, and it doesn't rain. Interesting. Right? And so the moral of that story is that a reliable antecedent is not necessarily a good predictor, right? You have to look at all the data, not just the data that ended with your event you're interested in. Why do you think it is then that these libid experiments have been, or the results of the libid experiments have been so reliably misinterpreted or conclusions were drawn that should not have been? I mean, you can speculate. One is, that, you know, it's a fascinating thing to think about, like, oh, maybe our decisions are

made, you know, unconsciously before we're aware of it. And you have this signal that very reliably comes before self-initiated movements. And it all just sort of adds up in a way, right? And if I think, you know, I'm always stumped at like, well, why did no one bother to ask? Like, we just assumed that the readiness potential was reflected, the outcome of a decision to initiate a movement. I'm not sure where we got that assumption other than, well, it does reliably precede the movement, but why no one really bothered to dig in and test that assumption? I don't know. You know, things like that happen in science. It's worth asking why, you know, regarding the readiness potential. It is, I guess it's true that the, the explanation

that I'm offering, the interpretation that I'm offering is a little bit hard. to wrap your head around than the simpler one it's i know some philosophers i've talked to don't like it when i call it a selectionist explanation but i think it is and and by selectionist i mean you know when when lamark in the in the debate between lamark and darwin when lamark saw giraffes and and saw that you know there were skeletons of animals that had you know longer necks and and, you know, getting longer over evolution, you know, over long stretches of time, right, over geologic time. He thought, oh, well, the, you know, the animal must have been trying, so something, there was,

there was like an effort here. This was intentional. The animal was trying to get its neck to be longer. Darwin's explanation is always a little harder to wrap your head around. It's like, no, it's just that if you take the one that's alive today, that's the one that survived and it survived just because it happened to have a longer neck and the ones that didn't died off. That's a selectionist explanation where say, look, if the conditions are met in whatever I'm looking at, like this could be the conditions for initiating a movement,

if those conditions are met, then it's kind of trivially true that if I look back, I'm going to see the antecedents that were necessary to bring it about. That doesn't mean that it was intentional. So, Professor, one of the reasons that I'm extremely excited to speak with you and have been since, well, for a little while now, is because you also studied consciousness. So at some point, I would like to get to what consciousness has to do with any of this, if anything. But prior to that, what I was going to ask, in the examples with the flu, the flu is still external. So do you have another analogy for people to, unless you're saying, that the conscious decision is also external. Is there some way that we can understand where

this decision is coming from? Right. Well, so the thought experiment about the flu is really just, I would say, what Daniel then it calls an intuition pump. It's imperfect. It doesn't match in every way. But it just is meant to give you the intuition that something can happen that looks like an early decision, an early, you know, would it be causal event, but that isn't. That, you know, with the flu, for example, you might see that it looked like your health was getting worse even before you came in contact with the virus. How could that be? Well, that's because we selected all the data where you, where the people came out with the, came down with the flu. And necessarily, you're, you're already a little worn down by the, you know,

when you come in contact with the virus, else you would have fought it off. But maybe it's easier if we just look at, that's a thought experiment, it doesn't match EEG data in every way. So let's just talk about brain data. And by the way, the buildup of neural activity before a spontaneous voluntary movement that you record with EEG,

which is recorded externally from the scalp, right? There have been experiments also recording neural activity invasively using electrodes in the brain. And those experiments tell very much the same story. So you have neurons in the supplementary motor area whose firing rate is just increasing slowly before the movement. And you have this in humans, in people being monitored for epilepsy, for epilepsy. see, and you have this in rats, you have it in monkeys. And oddly and interestingly enough, you also find it even in invertebrates like the crayfish. Does the crayfish have a readiness potential in their readiness discharge in their motor system?

Interesting. So just to say that this is not some oddity of EEG, right? It's a neural phenomenon, and you can record it in different ways. They all tell the same basic story. Okay, your stance is that the, the neuroscientific evidence is, in this particular case, to disprove free will is relying on faulty logic. But then that itself is not evidence for free will. You're just looking at this specific libid interpretation and then saying that it has historically been incorrect. It's been wrongly interpreted, I think. Yeah. And so the, the, the, The story that I'm trying to tell, the interpretation of those data that I'm offering is it doesn't prove that there is free will, but it just throws a wrench in the works of

this prevailing interpretation of that experiment that was used, you know, that was used to argue that we don't have free will. And as you rightly point out, it's not just that we don't have free will, but it's a little more specific in that, is that we don't have conscious free will, right? The idea that our conscious decisions actually are the determinants of our actions. Did you read Robert Sapolsky's book? Oh, yes, I did, yeah, indeed. I wrote it, read it.

So what did you make of it? It was a great read. Very well researched. I didn't agree with the overarching premise, but I thought he did a very good job of and a very thorough job of researching the relevant literature. And I think he was right. He talked about my work, my colleague Ori Miles's work, said, okay, you know, this kind of suggests that that interpretation was wrong, but that doesn't really matter. And I think he's kind of right about

that. But he talks about statistical determinants, you know, the fact that if you have challenging early life, then, you know, that predicts that later in life things aren't going to work out so well for you in various ways. You know, and there's lots of statistical data like that. But it's not, to me, it doesn't cancel completely the possibility that there's free will. To just to say, well, yeah, there are all these statistical reasons why where you're at now in life couldn't have been any different. And so basically everything's determined. I think he's right. We should take those statistics into account and we should give people some slack when, you know, because of their history. But that doesn't, that doesn't add up to, well, you don't have free will.

I don't think, right? I think, you know, if if there is something that would qualify as free will or as conscious free will, I think it's something that that happens relatively rarely. It's not like when you reach out to pick up your cup of coffee and take the next sip, that's not an act of free will. Most of the things you do on an everyday basis, I mean, that would almost be insulting to the idea of free will, right? So I think there's room for acts that generally count as acts of free will,

but they, you know, they might not be like your everyday actions. might be things that are relatively infrequent. What Sopolsky's talking about is just sort of like your general life situation, right? But that leaves plenty of room for acts, for acts that generally would qualify as free will. So I'm not saying that we do have free will, but I don't think that Sopolsky's argument, you know, completely renders it impossible or out of the question. I've been using Claude to prep for theories of everything for topics like, gauge theory or consciousness prior to interviewing someone like Roger Penrose. I need LLMs to engage

with the math and philosophy at that level. Now, that's just the online version of Claude. I also offline, for my personal use, used Claude Code daily. It runs in my terminal. It understands my code base. It handles extremely complex engineering tasks. In fact, you can see it here, a feature of theirs called Claude Artifacts that I used live in this conversation with Eva Miranda. It even impressed this professor of math. Oh my god, this is fantastic. There's no code that's required for that particular feature. I just described what I needed in real time and it built it instantly. Claude's research depth stands out. It synthesizes dozens of papers. It links unrelated ideas and surfaces philosophical assumptions inside some technical arguments. Ready to tackle bigger problems,

sign up for Claude Code today and get 50% off Claude Pro, which includes access to Claude Code, when using my link, clod.a.ai slash theories of everything. That's clod.com. A.I. slash theories of everything right now for 50% off of Clod Pro. That includes access to all of the features mentioned in today's episode. You know what's better than the one big thing? Two big things. Exactly. The new iPhone 17 Pro on TELUS is five-year rate plan price lock. Yep, it's the most powerful iPhone ever.

Plus, more peace of mind with your bill over five years. this is big get the new iPhone 17 pro at tellus.com slash iPhone 17 pro on select plans conditions and exclusions apply so tell us about noise the readiness potential and noise yeah because that's it that that is a key part of the of the interpretation that I'm offering if there isn't noise uh in evolved, then the explanation I'm offering doesn't really work. But there is. So that's, I mean, it's just well known. There's noise in the nervous system from microscopic levels all the way up to behavior, right? There's noise. And the important thing about that noise, and when we say noise,

I mean, what do we mean? It's just variability that we don't know how to account for, but that we don't. It's just if you've, all are the things being equal, if you just look at, you know, let's say neurons or populations of neurons firing, there's just variability over time that's not connected to anything that we can, you know, see in the immediate sensory context or whatever.

It's just so we call it noise. That noise, importantly, is what we call autocorrelated. And that's a key feature. Okay, and autocorrelation means... Yeah, so that's, it's, it's, it's both very important and, and, and, and, and, and, and, and, and, and, and, and, and, and, and, and, and also a little tricky, it's always a little bit tricky to explain auto-correlation. Um, it's, um, it's just sort of show somebody. I could show you some auto-correlated noise and say, see. And then, then it's immediate. You're like, oh, yeah, I see what, I see what you mean. You mean visually speaking? Visually speaking, yeah. Okay. Do you have slides or something that you can show them? Because we have the ability to show the audience. Oh, yeah, I sure do. I could send you a slide or, or I could send you a slide or. Oh, or. Oh, I could even share it right now. I don't know.

Yeah, why don't you share it right now? This way I can go through it as well. So that trace that you see on the top, that's auto-correlated noise. The one on the bottom is white noise. So now, you can just sort of see visually how they're different, sorry, and they're both noise.

They're both governed by a random process. Now, so let me sort of describe what you're seeing visually. and what's relevant about it. So with white noise, what's happening is that, by the way, going from left to right horizontally is time.

Okay. Right? So this is unfolding in time. And then up and down, amplitude, right? Let's say. Can you do me a favor and move your cursor? I want to see if I'm able to see the cursor as well.

Yes, I can. Okay. So if ever you need to point to something, feel free to use your cursor. If you need to. Great. I can do. Yeah, this lets me do this little red dot there.

so if you look down at the white noise here what characterizes white noise is that at every instant in time right the value that that this takes on is completely random and independent of the value that it took on in the instant just prior yes so so as this marches through time instant by instant one it's just it's it's completely independent from what happened before it's just random. So if you look at the autocorrelated noise, what you'll notice is that the faster fluctuations are smaller and the slower fluctuations are bigger. If you look at this, so you have, you know, this deflection here that I'm tracing out, right? This is a slower fluctuation, and its amplitude is much bigger than the amplitude of these tiny little fluctuations, right? So as the frequency gets higher, the amplitude gets smaller in auto-correlated noise. But what makes it autocorrelated is simply that the value that it takes on at the next instant in time is connected to the value it took in the instant just before. So it marches along and the change from one point to

the next is random, but it's the change, right? it's random, not the value itself. And so it's always, so it, it is always has some relationship to the value it had just before, in the, in the, in the, in the instant just before, right? So it's going to tend then to have where the, if it, if it climbs up like this, you know, quite a distance, it's necessarily going to do so relatively slowly.

Whereas with white noise, it can jump. anywhere, right? It can, it, it takes on it, it, you, you, you, you roll the dice again anew at each moment. There's no dependency in time. Let me ask you a question here. If you were to keep zooming out in time, would you then see that the top one would get farther and farther deviations or does it still have some tendency to be averaging near some middle point and it doesn't tend to spread much from that middle point because the white noise would stay yeah it'll be zero mean if it's white if it's Gaussian white noise then right right down oops right down the middle here it's just going to be zero and the mean right um i mean it depends on the process that that

generated the autocorrelated noise in this case it's also zero mean because i've just taken the white noise and forced it in a sense oh okay so i'm not asking about if the mean is going to remain the same because you can have the mean remain the same and your deviation from it also somehow remains the same like so but you can also have that that you just oscillate extremely over time so i just wanted to know if that was a property of the top one some knowledge that it needs to dampen once it reaches a certain point so it doesn't cross some threshold right it doesn't need to no again depending on the process that's generating it so one such process is called a drift diffusion process that doesn't that can drift as far as it wants or a random walk now the relevance here

of the upper of the upper auto correlated noise to the readiness potential is what right so let's let's get to that now so we see what this looks like the the point is that noise in the brain and that includes the motor system is autocorrelated we know that so it's not why if you see white noise in your EEG data, there's something wrong with your electrodes or your participant. But yeah, noise is auto-correlated, brain noise and neural noise is auto-correlated. And so if there's a threshold in the motor system that the activity has to reach in order to trigger the movement, let's say, to, you know, to, well, triggers a good word, right? there's a threshold and the noise right the the the underlying noise is autocorrelated if that

autocorrelated noise contributes in any way to crossing the threshold right what that means is that and i think i have it here here's a threshold to say there's this threshold right um the the dash blue line here yes um so what that means is that if you've reached the threshold it means that in all likelihood you climbed up there slowly because that's the nature of the noise the larger fluctuations are are slower ones okay and so like you see here right you know in order to reach the threshold and i've positioned the threshold so that it might reach it right it it it climbs up there slowly. And so if you were to align your data to these threshold crossings, and let's say you had many, many of them and average them, what you end up getting, and this is

what I've done is this very simulation, you get something that fits the shape of the readiness potential extremely well. Let me ask you a foolish question. Yeah. You keep saying that it climbs up slowly. However, what I see is that whenever there's a large jump, that actually happens quickly. So I have my cursor over it, but you're unable to see that. Let me see if I can describe it with my words. Okay, with my finger. The times where it goes up like so, it's quickly. Like maybe here, for example. Yeah, yeah, exactly. Yeah, I mean, it's so, so remember, this is all stochastic. It's statistical. So what, it just means that on, on average, the tendency is that the larger deflections

tend to be the slower ones, right? It doesn't mean that large, fast deflections can't happen, but there's a tendency for, you know, I think I see what you're saying. Okay, so let's just draw that line of the threshold that you have above, and that in order to get to that point from some lower point,

it will take you a longer amount of time because these little guys are fluctuating, slowly it's unlikely that you would have such a large fluctuation quickly that's right okay you're i got it you're gonna tend to climb up to it slowly and so how to the connecting that back to the readiness potential the idea is this okay that in the context of libid's experiment and any version of it. You are asked to move at, I mean, I don't know what the right word is, but at random times, move kind of whenever you feel like it. Don't plan in advance when you're going to move, right? These are the kinds of things that Libet asked his participants, right? It's a strange

task. Right. Right. So you're being asked to move for really, I mean, to put it in a simple way, for, yeah, for no real reason, for no particular reason. So my assumption was that, well, how, how, if I had to create a robot arm or a device that would, that could do libid's task, what are some ways that I might do that, right? And one way, I'm not saying this is the only way, but one way is just take advantage of noise inside the system lower the threshold so that it's kind of close to the noise right lower the threshold for triggering a movement and then just wait and sooner or later because the noise is is somewhat is random sooner or later it'll cross the threshold and you'll get a movement at a

random time problem solved right you've you've now you've yes you're you're you've complied with its instructions, right? Well, what I'm arguing is that that might be part of the way human brain complies with lipid's instructions, which is just to sort of either increase the noise floor or lower your threshold and weight, or at least that that's part of what's happening. So the noise is partly determining the precise moment at which that threshold is crossed. So when the experimenter is giving the instruction to the experimente, hey, arbitrarily wiggle your finger, you inside your brain are lowering this threshold. And so there's going to be some unpremeditated time that it will fluctuate. That seems to indicate that your decision and your action is still just based on something arbitrary, some noise at that point.

So, okay, that's a question I get very often. I should clarify that. In the computational model that I built to account for this, there are actually two important factors. One is the noise, which here you see it in front of you. The other is a slight, a slight, like, urgency or imperative to move sometime soon. That is to say that the participant in that experiment doesn't just sit there and just wait. there's what we call in in experimental psychology the the demand characteristics of the of that task the demand

characteristics if you haven't heard of that phrase it describes all that is goes unsaid right but that is still very real well what goes unsaid in that experiment is that yes i told you you know that that you can move kind of whenever you want to don't preplan it and so by whenever you want to technically you could wait till tomorrow technically right yes but you don't nobody ever does in every in any such experiment that i've ever run nobody waits longer than about 30 seconds at the longest typically it's around seven seconds on average nobody waits that all long right they know without another this goes unsaid that it's an experiment about movement and you're asking them to move at a random time, but to move, right? And so I think the subject knows, like,

you might be annoyed if they sit there for too long and don't move, right? So, so, so there is a slight push, a very gentle push towards the threshold that I include in the model, right? But that's a very weak kind of an imperative. It's there, and it's the, it reflects the demand characteristics of the task. And then superimposed on top of that gentle push is the noise. But the push is this imperative is weak enough that the fluctuations due to the noise are a little bit bigger. And so what that means is that the precise moment that the threshold is crossed depends a lot on the noise. So if you, I think I have a slide that shows that he, So you have, here's the threshold down here.

And if you see the red dotted line here, that red dotted line represents the imperative, the demand characteristics. So you know, you're kind of tending toward the threshold, but superimposed on top of that is all this noise. And so that the exact moment that the threshold is crossed is determined significantly by the noise.

And that exact moment, if the threshold is what then triggers the movement, that moment is, that's what you align your data to. And now you're going to look back in time from that moment on average. And what are you going to find where you're going to find exactly what you will always find in autocorrelated time series, right? If you back away from a peak, which is a slow exponential-looking climb up to that peak. So then what is going on consciously? Is consciousness noticing and reporting on decisions but not making them? Right. So that you have to qualify that question.

Do you mean in everyday life or do you mean in Libet's experiment? So let's start with Libet's experiment. In Libet's experiment, basically, yeah, you're asked to move without deciding consciously. That's literally what you're asked to do. Like, don't consciously decide, okay, I'm going to move now. so in that context in the context of of that experiment the conscious decision is the one you made to comply with the instructions right at the beginning of the experiment or at the beginning of each trial right and then you're like okay you know then i i've i've made this commitment to just comply with these instructions but in a way that i don't have to decide precisely you know at which moment i'm gonna move. I've just consciously decided I'm going to kind of let things play out so that a movement happens kind of soonish. So yes, in the context of Libet's experiment, that's exactly what you're

asked to do. Now, in everyday life, obviously, you know, it's very different from the, you know, narrowly constrained rules of Libet's experiment. When you say everyday life, you're not referring to grabbing a cup are you referring to deep moral decisions or something else i think there's a whole span but i guess yeah i'm thinking of you just movements that you make on an everyday basis one way one example that i use and this will be i guess more meaningful for american viewers but if anyone who's familiar with baseball if you know if if you're trying to hit a baseball that's throwing at you know at high speed right um you you you really don't want noise involved in that decision you want the the the the stimuli in front of you to drive that decision right and and the

baseball or the body the the the movement of the pitcher so on that should drive but what happens you know when kids learn to play baseball we start them out in something called t-ball where you you put the ball on a little stick. Yes. It's not thrown at them. There's a stick that's about one meter tall. You set the ball on top of that stick and then the child can just hit the ball. So if that's the context, it's a T-ball, right?

And nobody's telling you when to swing the bat, then what determines, not the moment when you swing the bat, but the precise moment when you swing the bat, right? what determines that what what determines whether you swing the bat like right now versus 200 milliseconds earlier or later well that is a context where some of you know some internal noise could potentially play a role and in in you know determining the precise moment which you swing the bat and so there's this now you can think of sort of a continuum from the t-ball to the baseball yes right where for many kinds of actions, you don't want noise to play any role, but for actions toward the other end of the spectrum, maybe it does. So that answers the idea. People say, oh, well, does that mean that all my actions that I do every day are just random? Well, obviously, no. But in context that resemble

the libid experiment, like where something like T-ball is a little bit closer, then yes, maybe. so you mentioned 150 milliseconds i believe that's the number you mentioned earlier yeah okay i want to talk about that because even if the readiness potential is noise and there's some threshold crossing that counts as a decision and it still precedes reported conscious intention by a small amount which is 150 milliseconds much shorter than the one sec approximately one second i'd mentioned earlier but it still precedes it so so help me understand understand the 150 yeah so no the 150 milliseconds i'm referring to is the one is 100 50 milliseconds before the onset of the movement not before the decision uh um so now if you use libid's method to ask people

when they felt the decision or the urge to move that clocks in typically at around for libid it was around 200 when i've tried to replicate it it's been around 150 it's around that moment in time. So what this explanation offers is an answer to what I call Libet's paradox. Libet's paradox is, well, wait a minute, if they say they moved at, you know, if they say they decided at 200 milliseconds before the movement, how is it that the brain seemed to be building up much earlier? And the answer to the, my answer at least, that I offer to the paradox, to that paradox is that in fact the decision happens that the actual triggering of the movement and the motor system happens at about 150 milliseconds and guess what that's just about when people declare

feeling that they reporting that they felt the urge or the decision to move the conscious decision so so the conscious decision actually aligns pretty well with where I think the actual commitment or trigger of the movement happens. And so maybe, you know, Libitt's error was to just not take his subjects at their word, right? He thought, oh, well, there's this readiness potential. So my subjects must be wrong.

Right? Their movement's beginning far earlier and it's beginning unconsciously. So why are they mistaken? Why are my, you know, and maybe the answer is, well, they weren't mistaken. Maybe they were giving you a very approximate guess,

or estimate as to when that decision happened. And then that wasn't, that, that, that guess wasn't too far, too far from the, from the, from the reality. Okay. My understanding is that it's a mistake people make when they say that Libet didn't believe in free will, that Libet did, but he changed it to free won't, namely that you have a veto power. So you can decide not to do something, but that your body is initiating you to do it.

And you can say, no, please. That's right. he advocated for this idea that, well, we're not in charge of initiating our movements, but we can veto them. That was an idea that he had. And the idea was that those last 200 milliseconds. So in his experiments, roughly speaking, the decision clocked in about 200 milliseconds before the movement, the conscious decision. So he thought, well, that gives you 200 milliseconds. to then intervene and inhibit or veto the movement. And that that is something that you can do voluntarily, consciously.

And your views on this are? So it's rare that I would say something that's strong, but I think he was just absolutely wrong. So, you know, we have newer data now showing that 200 milliseconds is actually roughly where the point of no return is. Meaning that after 200 milliseconds is precisely when you can no longer inhibit your movement. So if you look back from self-initiated movements and you see these experiments have tended to be where, for example, you just interrupt people at random times and say if you get interrupted, then stop, stop what you're doing. Okay.

Right. And if you, you know, these interruptions come at random times. And if they happen to come within that last 200 milliseconds before the movement, the movement, the movement. movement just happens. So what's the difference between that and just reaction time? Um, it, it's because reaction time is, is faster. My understanding is it can get up to 50 milliseconds. So it's not a, it's a, it's a, it's different from reaction time task, because you're,

you're kind of doing something like Libet's task. So you're, you're waiting to move at a random time. Then you get an interruption. And if the interruption happens when you were very close to executing a movement spontaneously, you're not able to inhibit the movement anymore. It just, the movement happens. Can you spell it out extremely clearly with an example, like the experiment he is told to play ping pong or they're told to walk and stuff?

Spell out exactly what's happening. Well, let's say it's, let's say it's just Libby's experiment. You're asked, I mean, that's where this has been. done anyway, you're asked to just either press a button or flick your wrist, make some sort of an abrupt movement, again, at an unspecified time that should be not pre-planned by you and that should just be sort of spontaneous, right? So you're there getting ready to move. And now the clock, you know, let's say slow down time, right? So now it's, now let's say, and imagine that we know the future. So we know that a movement's coming in 500 milliseconds. Okay.

So 500, 400, 300, 200, is getting closer closer. And now we cross the line of 200, and now you hear the beep that means don't move. Guess what happens? You move because you were only less, you were less than 200 milliseconds from making that spontaneous movement. It's too late. It's already, the process is already, you might call it ballistic. Yes. And so you can't interrupt, it's too late. You can't interrupt it now. So 200 milliseconds is a pretty good candidate. And I think the data are pretty clear that that's roughly the point of no return. So that can, it's either the point of no return or it's the point when you get the chance to veto. It can't be both, right? They're in absolute opposition to each other. And so I think a little bit was just wrong. The 200 millisecond mark is,

is the opposite of the time after which you can inhibit your movement. It's the opposite of that. It's the time after which you can't inhibit your movement anymore. Interesting. Now, speaking of ballistic, my understanding is that there's a linear ballistic accumulator model, which explains the readiness potential just as well without stochastic fluctuations. Am I incorrect? You're incomplete.

There is randomness in the linear ballistic accumulator model. Basically, what that model does is it takes the randomness that evolves over time in the epoch and then pushes all that randomness to the very beginning of the trial. And so basically the random thing now is just the starting point, where you start at. And so obviously, if you start closer to the threshold, you'll hit it more likely and earlier. And if you start farther from it, it'll take longer. So from the moment the trial starts, after that, it's just, it's, it, there's no noise. But the starting point is random. And, and in my mind, that's more or less equivalent to just taking the noise as it unfolds,

or would have unfolded over time and just pushing it all back and, you know, pushing it all back to the beginning of the trial. So, yeah, it can, it can fit the shape of the readiness. but it doesn't do away with randomness. There's randomness in that thing. It's just at the very beginning, right? And it's, in my mind, it's highly unlikely that that is, is the way things work in the brain. It's what I would call a toy model.

It's good for as a thought experiment. Okay, thank you for correcting me. Yeah, I mean, I don't think it gets around the problem of randomness. So, well, thank you for that, because I, I didn't know that. And why don't you tell me, or tell us the audience as well, what are some of the objections that you get from your colleagues? Um, I mean, one of the more common ones, and that I think I already tried to address was the one, oh, well, does that mean that, you know, my movements throughout the day are all just determined by noise? And that's like, that's an unfair caricature of the model.

And I think I explained why. they're not determined by noise but then when the movements are more unconstrained then noise may play somewhat of a role right so that's one that i that i i hear from time to time you know to be honest when i was going to you know when i was getting ready to publish this i expected it to be hard to publish because it kind of went against the grain went against that prevailing view and And it was. It was a bit challenging to get it published. But then the reception I found was a lot of people came to me and said, you know, I always thought there was just something not quite right about that whole story. And so I'm glad you came up with this alternative explanation. Right. So while I've gotten some pushback from some colleagues, a lot of colleagues have been really positive about it, even if they, Even if they're skeptical, like, well, maybe, maybe not, but they like having the idea out there because it's another plausible explanation. Tell me about your favorite consciousness theory. Ah, consciousness. Yeah, and why, like, why these are even, you know, related consciousness and free will. I would say, I don't, do I have a favorite theory?

It's just that there are different kinds of theories. I think there are two things. One is explaining the mechanics of how consciousness works in the brain. And I think why some patterns of brain activity evolving over short bits of time are consistent with that information being accessible to you consciously.

I think that some kind of a hybrid between higher order thought theories, global workspace theory, and integrated information theory, I think those each touch on necessary conditions for consciousness in the brain. They each have something to contribute. And I like the idea of a hybrid between those three.

But then there's the bigger question of explaining not just the mechanics of how consciousness works in the brain, But trying to get at that hard problem of consciousness. How is it that something that is built up from matter can be responsible for a subjective experience, right, for something that it just seems very mysterious, right? when you think about a conscious sensation, like just seeing a vibrant color or smelling a very pleasant smell, how does that come from workings in the brain?

And I think the theory in my mind that I favor now for addressing that is attention schema theory, is Mike Gratziano's theory. Okay, so there are four theories here that have been mentioned. Higher order, global, IIT, and attention schema.

So the hybrid that you are supposing will be the actual theory is between these four or between the initial three? So the hybrid that I think will account for just the mechanics of consciousness in the brain

are the global workspace theory and higher order to thought theory and IIT. Those three might have some things to say about the mechanics in the brain. Attention schema theory is by itself I think the answer to the hard problem.

And that's where I'm pretty lonely these days in the field. I think there are, I think there are more and more. There are people who are either coming on board with that kind of a theory or at least being more open to it. Yeah, I'd like to speak to Michael Graziano on the podcast. I haven't yet. Oh, yeah. So he would be a great person to talk to.

Yeah, I quite like him. his theory. When I first read, I'll, you know, how sometimes you have an event in life where you just, you never forget it, you never forget, like, where you were and what was going on at the moment that something happened. Yes. Right. And I, I was riding, I was riding on the metro in, I was riding on the commuter train, I think, near Paris. And I was reading Gratziano's 2011 paper where he introduced this theory. And I, to be honest, I, I hadn't known about it yet. And I was reading that theory and I was getting most of the way through it and suddenly it just hit me like a ton of bricks and I was like, whoa, this, this is different. And, and this, and, and it kind of, I mean, I was,

his theory is very much in line with, with Dan Dennett's philosophy. You know, he, Dan Dennett offered this same kind of idea. He just didn't, he didn't offer like a brain mechanism or, or, you know, why this happens. Well, that's what, you know, that's sort of what Graziano offered. He said, well, you know, you can have this idea that, that, you know, the question we need to be asking is, why are we so convinced that there's something else in our brains other than matter? Even though if you go dig around in the brain, there's nothing else in there just matter. But why are we so convinced that there's something else? Gratiano came up with an explanation for why we might come to that conclusion and be really, really vehemently stuck to it. And I think that's the beauty

of the theory. And so the question is, do I, do I need to explain consciousness? What if I just explain why, if I, or do I need to explain the hard problem? Or is it enough to explain why we are so convinced that there's a hard problem? Interesting. What did that do to your sense of identity? Not much, really. it didn't really threaten my sense of identity in any way. I think that that consciousness is still is what it is and what I, you know, the things that I perceive are as real as they ever were and as beautiful as they ever were. It's just that there doesn't have to be, you know, I think that what's

implicit in the hard problem in my mind is the idea that there's you know there's all this there's matter there's neurons and glia and all what you find in the brain and then there's consciousness which is something else and you know attention schema theory the idea is that the something else just evaporates it's just why are we so convinced that there's something else you know gratziana has this beautiful story of a neurologist And his patient, and the patient says, you know, there's a squirrel in my head. And this person had sustained brain damage, I think a stroke through a stroke. And the neurologist says, well, you mean, you know, you have the thought of a squirrel and you just can't get it out of your head.

And the patient says, no, no, no, no. I actually have a furry rodent living inside my cranium. 100%. Like, he's in there. And the neurologist says, well, look, I can show you your brain scan. and there's no squirrel in there. And the patient says, well, I know it seems irrational and it doesn't seem like it could be possible.

And yet, I swear to you 100%, I'm sure of it, there's a squirrel in my cranium. And so, you know, the question is, I always joke about this when I give talks, it's like, well, should I go, if I encountered this patient, should I go and try and get a grant? to search for the squirrel? Or should I go and try to get a grant

to figure out why this guy is so convinced that there's a squirrel in his head? And the idea is that, well, maybe consciousness is the same, right? Maybe what we really need to understand is why we're so convinced that there's something else in there. When great thinkers going back a very long time have adamantly said,

there seems like there is nothing else in there. Right? They all come to that, you know, like even if you go back to Leibniz, he has his mill analogy. Like, if you could walk into the brain as if it were a mill, you would just see a lot of mechanical things, gears pressing against one another, but nothing that could explain a perception, right? That could explain what it's like to experience the smell of a rose or, you know, etc. So I think that, you know, whether you like the theory or not, it sure raises a really good question, right? You know, if I've explained why you're convinced that there's something else, am I done? Do I need to explain anything else? Check out the big stars, big series, and blockbuster movies. Streaming on Paramount Plus. Cue the music.

Like NCIS, Tony, and Ziva. to make up our own rules. Tulsa King. We want to take out the competition. The substance. This balance is not working. And the naked gun. That was awesome.

Now that's a mountain of entertainment. So where does the What It's Like come from? I mean, for one thing, the phrase what it's like, already is interesting and informative, right? It suggests relations, right? You say what it's like. Well, it's like

something else, right? And we know that our perceptual brains and our memories are associative, right? The way we remember things is in relation to other things. So

when we say there's something it is like, it's it's it's saying there's some association right and and and and that there's something else that it's not like so that this particular experience this particular information this sensory information that I'm processing now is embedded within a network of associative relations and it's like something right and unlike other things now why is it that you might say well why is it there's something it is like at all in the first place what attention schema theory says is that information that is modeled by the attention schema is and and this is something that you know

we don't know the answer to yet but it proposes that it's it's sort of flagged in some way so that when asked about it, we can answer the question directly and we can say things. We can report about it in that way. It's flagged as being different or special, different from information that's encoded outside of your awareness in that moment.

So let's say we can make formal analogies and say that Jupiter is like Neptune because they're both spherical but they're different in that they're at different radiuses away from the sun or distances away from the sun but in that they are like scenario there's nothing about a feeling now there's a feeling of being tormented right okay so i don't even need to say it's as if or it's like that in that in that scenario i could just say i feel that i'm tormented or feel suffering or what have you so So this feeling. Pain is often a good example.

Yes, yes, yes, because it's undeniable, and it's something that can always be turned to the person who is the illimitivist or the person who thinks consciousness is an illusion and say, well, say that to the prisoners of Auschwitz. So explain this feeling, maybe not the feels like, because the like can be then said it's an analogy, but just the feeling. Right. And I think that's the, I mean, that is what philosophers and scientists appeal to when they talk about the hard problem, right? It's the feeling. It's the what it's like in the first place, like the root level, what it's like. I think, so here's where I'll, I'll admit that just I'm a firm believer that it's better to walk around thinking you're ignorant. and end up later being right, then the opposite. That's to walk around thinking you've got the answer and being wrong. And so here, I don't know. I don't think we know yet. But my hunch is that we can use an explanation or a theory like attention schema theory

to get at that question. And that, yes, that explanation will, kind of dissolve the question in a way. And one of the nice things about attention schema theory is that it predicts that it predicts why people won't like the theory. Because it's just not satisfying, right? It doesn't, that question you just asked, like, you know, if I said, when I'm experiencing pain, it just hurts, right? What is that? How can I, how can this possibly account for that. If attention schema theory were to say, well, I can explain all the things that you're saying right now and doing. I'm going to explain all your behaviors right now in

terms of brain activity. And you might say, well, yeah, those are just my behaviors. Those are the things I'm saying, but there's this undeniable thing that I'm feeling. Now, if the explanation is complete from an objective point of view, you may still feel dissatisfied with that explanation. Like, well, no, there's something missing, right? So I think that's one of the things that is neat about. The theory is that it predicts that you won't be satisfied with it. We never will because we embody, it's the very same brain that is trying to explain the hard problem is the one that's kind of, if you take illusionism to be the right answer,

it's the same one that's plagued by it. There's a concept called coercive theories of meaning, and it's from, I believe it was pioneered by this guy named, or woman, I apologize, if I just know the last name, it's pigeon. And I'm not even entirely sure if that's the person who first proposed this. But the idea is that many times in philosophy, when it comes to meaning questions, Can dogs fly?

You can say, well, that's a meaningful question because we can test it. But then from another point of view, it's unmeaningful because all that is meaningful is are well-formed sentences in a first-order language, and this doesn't fit that, and you would have to predicate over something, you'd have the scope over or something in this real world. Now, few people take that latter theory of meaning, but the point is that there are questions that you can say are meaningful under some theory of meaning. meaning, and then you could just say, well, it's unmeaningful. It's nonsense in some other point of view. And so it's often the case that when one takes a certain theory of meaning, many questions

you would want to know the meaning of are then called, that's just not meaningful. And then a host of questions that you think should have no meaning, all of a sudden do. So, for instance, in naive set theory, it's the set of all sets is, is seen to be paradoxical. And then when you have ZFC, then all of a sudden you have other paradoxes that come in like the Lowenheim-Scholen paradox, or it's not actually a paradox, but it's just an unintuitive result, or the Benak-Tarsky paradox, which is an unintuitive result, not a paradox. That one has to now accept. So it seems like there's always a problem with any given theory of meaning, that from someone else's point of view, it would just be, you're just declaring a certain theory of meaning, and then saying that my question, lies outside of that, but I could equally take some other theory of meaning and say that your questions lie outside of this. And now we're just fighting over theories of meaning. It's not like we're getting at the question. So sometimes when someone says, we've dissolved the hard problem, to the person who believes there's a hard problem, it seems like you're just dodging the question.

And then I'm sure the other point of, the other person has the other point of view. Like, you're just wrestling with something that's meaningless. You're trying to fight, you're fighting fighting a cloud. Right. I think that's why they call it the hard problem. But it's hard for a different, in a sense, seeing it from an illusionist perspective, which is Graziano's theory is in that category.

That's the nature of the hard problem. It's not hard in the, I think, philosophically, what would that be, the metaphysical sense? It's hard in the epistemological sense. it's for us right it's a hard problem so that let's just say right let's just say that this theory that's mechanistic is correct and it's accounted for everything and that you present it with this individual who insists you know who you know who like all of us me included feels there's meaning

here that's not being accounted for um that correct theory is not going to be satisfying to that person and that may be where it stops, right? It just, okay, that's it. I mean, we, if we start to agree that that is the right, it's a complete explanation, we're going to be stuck with feeling dissatisfied with it at the same time. And we can't go farther than that, at least could be one way that that goes.

You know, the word, I, you know, we, there's a lot to be, you know, I don't think we're as far along in our understanding of consciousness as a lot of people think. So, you know, time may tell. Tell me how your work has evolved over time. Speaking of time, yeah.

You were initially interested in consciousness and perception and you did some work on the libid experiments' interpretations and you conducted your own experiments, so you went into experimental neuroscience as well. Walk me through your journey. I did. I started out being interested in consciousness and perception and attention. So those with, and I, as a PhD student, that was what I worked on. And then as a postdoc in France,

working with Stand A Han, I worked, I continued, I was continuing to work on questions about consciousness and perception, threshold level perception. But it was at that time that I started having these sort of ideas about the libid experiment. And then I kind of concocted the experiments that I ended up doing while I was in the middle of doing another study. And I went to stand and I said, look, I know I'm like in the middle of doing this other study, but I have this great idea and I want to just run with it. And, you know, I said, can I explain it to you and then, you know, tell me what you think because it would mean kind of slowing down on the other project so I could get in there and get so I explained it to him and he liked the idea and he was like okay do it and

that became that work that I did that I published in 2012 on the libid experiment so that kind of came from this you know from outside of of what I was what focusing on or had been focusing on up to that point. I did go back and finish the other work as well. And I'm coming back to that now, we're looking at stability in networks of neurons as a hallmark of conscious processing in the brain. So I'm still working on that. Interesting. The stability of neurons as a hallmark of consciousness. The stability of neurons or a stability of certain type of activity. Yeah, okay. Of the activity, yeah, that populations of neurons and the activity within populations of neurons has to be transiently stable for a little bit of time, maybe 100 or 200 milliseconds,

in order for the information that they're carrying to be experienced consciously. So a necessary condition, not a sufficient one? It's a necessary condition. It's not a sufficient one, yeah. It's a necessary condition. but it's part of a broader kind of model of perception and perceptual decision making but that that I think is also relevant for for consciousness so I'm still doing work on that and it nicely I published that work in in 2015 about stability and then didn't hear much about that topic for a while and then just I think it was last year 2024 this beautiful paper came out by someone named I forget his first named Eisen his last name and working in Miller's lab looked showing that indeed stability was a nice signature of conscious states and that things were unstable when people were in an anesthetized state so for me that sort of reawakened a bit my interest in that topic I think there's a lot explanatory power potentially there I mean one of the ways you asked I guess I should come back to your question one of the ways

that my, you know, thinking has evolved is sort of my understanding of how consciousness and questions about free will are related. Interesting. Because I've worked on, I've worked on both. And, you know, I think some people who are skeptical about both of those kinds of research, and I'd say, oh, well, look, you know, consciousness is very mysterious, it's kind of fringe, and, you know, free will is also mysterious and it's kind of fringe, so somehow they go together. And I think that caricature is, well, it's missing some very, very important

and meaningful deep connections that are there. So for one thing, and you brought this up yourself, you know, if you ask someone who's not trained in philosophy, just ordinary citizen, ordinary, thoughtful person, you know, what they think about free will. By and large, people want to know that their conscious decisions drive their actions, right? That's when then that's what they're thinking of when they think of free will. I consciously decide something. That conscious decision drives my actions, right? So there's an a priori reason why we would want to, you know, have, why we see a connection between consciousness and free will, because the kind of free will that most people think about is one where conscious decisions, then drive actions. So consciousness is, is important

a priori but there's a I think a deeper reason why I think those that initiating action and consciousness are necessarily related and that comes back to I had these ideas initially again when I was a PhD student and I was working with blindsight patient whose initials G. Y, he was one of the more well-studied blindsight patients. Now, blind sight is when, you know, you've had damage to your primary visual cortex, and that part of your visual field is now blind, but we can present stimuli in the blind part of your visual field, and the patient can guess, if forced to guess, they give very accurate guesses, but they don't know why they're getting them right. Yes. That's blind sight. And one of the things that struck me,

when working with a GY and when learning about blind sight in general is that if you ask people very often if I stand in the blind hemifield of the blind sight patient so he can't see me and I toss a tennis ball to him what happens right and a lot of people say oh he just catches the ball right and and that's

absolutely wrong no he just nothing the ball just bounces off of his shoulder he's like, why'd you throw the ball at me? So, okay. Right, a lot of people are confusing that with, like, with Parkinson's. So in Parkinson's, someone who has a really hard time initiating movement, they can still react to things. But here we're talking about blind sight.

And so in blind sight, one of the, these, it's a little detail, but boy, is it an important one. When you do an experiment with a blind sight patient, yes, they guess correctly, but you have to tell them when to guess. And that instruction, you're telling them when to guess, has to be something that you deliver consciously. So it can be like a beep or it can be something in the part of their visual field where they still have sight. But you have to cue them and say, okay, ready? Okay, now, make your guess.

Yes. And you know, about something. Like if you show them an oriented rectangle that could be horizontal or vertical and you say you have to guess whether it's horizontal or vertical, you have to tell them when to guess. If you don't tell them when to guess, if you just say, you know, when the rectangle appears on the screen that you can't see. Yes, okay, I see. Press the button indicating its orientation. The patient will just sit there. They'll do nothing, right?

And you don't have to, you don't need this blind sight context to see the absurdity of it. Imagine if I were to tell you, raise your hand when the invisible stimulus appears. Will you ever raise your hand? right if you did raise your if you did raise your hand what would you then tell me about that right you'd say oh i just raised my hand but i don't know why yes so that's absurd and there's a reason why it's absurd is because it doesn't happen right it my theory is that consciousness is necessary for initiating a new movement you can you can guide an existing or ongoing movement or sequence of movements with unconscious information,

but you can't trigger or initiate a new movement, right, without being conscious of the stimulus that triggered it. And now, when I say necessity, it's logical necessity. I don't mean temporal precedence. It doesn't mean that you have to be conscious beforehand, but that when you look at the episode as a whole, because some people say, oh, well, how could you hit a baseball, right? If you wait till you're conscious of the flight trajectory of the baseball,

you'll never hit it, right? True, true. But what I mean is logical necessity. I mean that in that whole episode of time where baseball is thrown, you swing the bat, and maybe you hit it, if I ask you, why did you swing the baseball? Why did you swing the bat? You're going to know.

You're going to be conscious of the reason, because the pitcher through the ball. You don't have instances where you make a movement in response to a stimulus, a new movement that's initiated, triggered by that stimulus when the stimulus is processed unconsciously. And what's nice about that is that it's disprovable. So you could disprove it. Do you think we could neuroscientifically disprove free will? Yeah, I mean, I think we could disprove free will.

I think it's possible to disprove some sort of more naive conceptions of free will. I think that we could disprove conscious free will. Let's be clear. We haven't done so yet. And maybe we never will. I don't know. But it's possible. So I think, and it's a good question, what would it take? What would be the convincing evidence, right? My colleagues and I recently wrote an article in Scientific American about it. this on the topic of where should we set the bar for evidence against free will and i think we should set the bar a little bit high but i think in order to disprove conscious free will you'd have to be able to predict my actions well in advance so many seconds at roughly a hundred percent accuracy and not just mundane actions like there's food in front of me and i eat it right but and I'm hungry. There's food in front of me. I'm hungry and I eat it. Okay, great. But it has to predict

sort of every action, right? So you'd have to come up with context in the lab where the actions were not too routine. And if you could do it reliably close to 100% correct, several seconds in advance, that would give us something to talk about. Yes, if it was food in front of me and I'm hungry, I'm eating it 100% of the time. I love chips, especially potato chips. Oh, my gosh. Exactly. It's not much of a predictive feat if we do that. So tell me, what's exciting you these days research-wise?

Oh, yeah. So I'll start with one that I consider kind of a fun experiment and a fun possibility. We're looking at, and I have a few experiments going on looking at this question. So, you know, it's been forever. I mean, as long as the modern scientific study of consciousness, has been around, people have been debating whether normally cited subjects in a visual experiment, when you get them to where, you know, they're like 75% correct. So they're definitely way better than a random guess, but they're not, they're far worse than, you know, than

clearly seeing them, right? Okay. So the question is, you know, if the subject says, gee, I really felt like I was guessing. I'm not sure I was conscious of anything. Right. Do you take them at their word? Were they really not conscious of anything? Maybe they were just reluctant to say that they were. That argument's been going on forever. And one idea, one hypothesis that's out there, or let's say a camp, right? A theoretical view is that the threshold for being conscious of something is higher and steeper than the threshold just for being able to perform the task. And that in between there, you have something like blindsight where you're guessing and you don't really know the basis upon which you're making those guesses. So we wanted to test that. We've been trying to test that in a couple of different ways. One way that we're looking at right now is we're measuring people's skin conductance while they do a threshold level of visual discrimination task. where you know things range from being you know down to just a to what we call chance level

meaning they really are guessing all the way up to you know they're getting them all right and then everything in between and what we're interested in so the skin conductance is a good measure of surprise so when you're surprised by something you you get an increase in skin conductance and and that surprise doesn't have to be like something at a conscious level right it can, and it can just, it's like a gut level, uh, surprise. It's something unusual. Um, so what we do is we have them do blocks of, of trials of this task. And at the end of each block, we show them their score on the screen. And we look at their skin conductance in response to seeing their own score. And the, the premise is, the prediction is that, well, if, if it's true that the

threshold for consciousness is higher and steeper, then when their scores are kind of in the middle, like, you know, between, let's say, 65 and 85% correct. Yeah. And they know that getting that kind of a score is very unlikely by accident, that they might be surprised by their own score. If they were genuinely not conscious of the stimuli, then they might show this autonomic surprise response to seeing their own score. Whereas if their score was really close to chance or close to perfect, they'll be less surprised. Because obviously, if you get them all right because they were clearly visible, you're not going to be surprised that your score was 100%. And likewise, if they were nearly invisible, you're not going to be surprised that your score was bad, was close

to a coin toss. But in the middle, there's the question. Are you surprised? Are you surprised? You're not by your own score. So we're doing that. That's an experiment we've got going on right now, and I'm really curious to see if we see signs of surprise in that middle range. Well, Professor, I'm honored to have spoken with you for this long, and I hope to have you on again, especially to do a deep dive into consciousness in these later experiments,

and hopefully there's something surprising about the experiments, about surprise as that's always great to that's an indicator of learning so hopefully i can speak to you again about that and the audience i'm sure has appreciated your ability to make extremely abstruse topics clear and i'm sure they have plenty of questions so if you're listening slash watching please leave questions that you have for aaron in the future and the professor, we'll get to it when we speak again. Happy to do that, yeah. And it's been a delight to be on your show.

And I'm also honored to have been on your show. And I wanted to also just congratulate you on the success of your show. It's excellent. I'm also a fan. That's great. Thank you so much. Hi there. Kurt here.

If you'd like more content from theories of everything and the very best listening experience, then be sure to check out my substack. at Kurtjymungal.org. Some of the top perks are that every week you get brand new episodes ahead of time. You also get bonus written content exclusively for our members. That's C-U-R-T-J-A-I-M-U-N-G-A-L.org. You can also just search my name and the word substack on Google. Since I started that substack, it somehow already became number two in the science category.

Now, substack for those who are unfamiliar is like a newsletter, one that's beautifully formatted, there's zero spam, this is the best place to follow the content of this channel that isn't anywhere else. It's not on YouTube, it's not on Patreon, it's exclusive to the substack, it's free, there are ways for you to support me on substack if you want, and you'll get special bonuses if you do. Several people ask me like, hey, Kurt, you've spoken to so many people in the fields of theoretical physics, of philosophy, of consciousness. What are your thoughts, man? Well, while I remain impartial in interviews, this substack is a way to peer into my present deliberations on these topics. And it's the perfect way to support me directly.

Kurtjymongle.org or search Kurtjymongle substack on Google. Oh, and I've received several messages, emails and comments from professors and researchers saying that they recommend theories of everything to their students. That's fantastic. If you're a professor or a lecturer or what have you and there's a particular standout episode that students can benefit from or your friends, please do share.

And of course, a huge thank you to our advertising sponsor, The Economist. Visit Economist.com slash tow, T-O-E to get a massive discount on their annual subscription. I subscribe to the Economist and you'll love it as well. To is actually the only podcast that they currently partner with. So it's a huge honor for me and for you, you're getting an exclusive discount. That's Economist.com slash tow, T-O-E. And finally, you should know this podcast is on iTunes, it's on Spotify, it's on all the audio platforms. All you have to do is type

in theories of everything and you'll find it. I know my last name is Complicated. so maybe you don't want to type in gymongle, but you can type in theories of everything, and you'll find it. Personally, I gain from re-watching lectures and podcasts. I also read in the comment that toe listeners also gain from replaying, so how about instead you re-listen on one of those platforms, like iTunes, Spotify, Google Podcasts.

Whatever podcast catcher you use, I'm there with you. Thank you for listening.