Hacker News Reader

Comment by n4r9 | Hacker News Reader

This looks like it's related to the "Orchestrated objective reduction" theory of consciousness [0], which is a brainchild of physicist Roger Penrose and an anesthesiologist named Stuart Hameroff. After 30 years it continues to have very serious problems and is generally rejected by physicists, mathematicians, computer scientists, and philosophers.

[0] https://en.wikipedia.org/wiki/Orchestrated_objective_reducti...

View on HN

Rejected by other experts who also have no idea how to explain consciousness.

Your theory having serious problems but no competitors does not actually solve the serious problems with it.

If I can't tell you why the sky is blue, it doesn't make your theory that it's green more likely to be right.

Funny example. There are languages where sky is kinda green.

Vietnamese: The word "xanh" can refer to both blue and green.

Japanese: Historically, "ao" (青) could refer to both blue and green.

Welsh: "Glas" can mean blue, green, or gray

We're talking about people like Marvin Minsky or Hilary Putnam, who have made very significant contributions to the discourse. And if Max Tegmark thinks your claims are a bit too far out, you've got your work cut out.

Marvin Minsky, Tegmark and Putnam have nothing to do with neuroscience, and have no authority to speak about the nature of consciousness.

Why neuroscience has a monopoly on Consciousness?

Because it studies the only known vehicle of consciousness - neurons and their networks.

It's not known. It's a belief hold among some scientists.

It also assumes materialisam is true.

Ahaha. No, I am a neural network, and I am conscious. Destroying my network will destroy my consciousness.

Not necessarily, there are plenty of rigorously documented cases of people being conscious without any brain activity.

This article is also pretty interesting: https://www.frontiersin.org/articles/10.3389/fpsyg.2022.9555...

I am not aware of rigorously documented cases of people being conscious without any brain activity. Can you point me to some?

Are you aware of any documented cases of observing brain activity without using consciousness?

Yes- that's a widely done thing. In particular, brain activity is evaluated using EEG, MRI, reflex response, and many other physiological methods.

Consciousness is a primary requirement to see, operate, read results from all of that.

Stop trolling.

Is that how your llm model reacts when you can't formulate back a reasonable argument?

It’s not trolling, it’s an obtuse reference to the unsolved measurement problem of Quantum Mechanics.

If that's equivocation as a wisecrack, it's particularly unhelpful in discussion about a pseudodiscipline whose proponents do no small part of their bullshitting by conflating the everyday psychological sense of "observation" with the sense of physical measurement.

Physicals measurements like everything else can be read only through consciousness. I was not talking in terms of wave function collapse. Just simple fact that consciousness is necessary to know there are experiments happening. Consciousness is necessary for you to experience the world and arrive at this conclusions. You take consciousness for granted.

See linked article, under "Phenomenon #6: Cognitive abilities can be retained when the brain is seriously compromised", terminal lucidity, etc: "the patients demonstrated normal cognitive abilities just prior to death, contrary to what objective medical findings would have predicted (e.g., EEG, neuroimaging). These patients are operating in an anomalous manner that brings into question the idea that the body is a “puppet” controlled from the inside (the brain) and that perhaps it can function alternately in some instances."

I'm sorry, I can't take anything in that paper seriously, its references are "case reports", which are not rigorous documentation. They're observations mixed with narrative (even the one reference they say is prospective does not appear to be). This is not science, and there are explanations/reasonable alternatives for all the references that make more sense. In short, I question that claim "the patients acted in ways that would have not been predicted by EEG".

Without brain activity, all that can be demonstrated is the recollection of being conscious. It is well known that we can recall lots of things that never happened, with our brains filling in the blanks with invented details.

> It also assumes materialisam is true.

As it should until we have a better theory we can test.

Looking at neurons doesn’t explain the plane of consciousness.

So? That doesn't rebut the comment you are replying to, which basically states that currently our only window at all into consciousness is through neurons and their networks.

At least neuroscientists are actually looking at the only physical medium we know of that can hold consciousness. Everyone else is just basically philosophizing, at least until anyone can show some other medium capable of holding consciousness.

Our understanding of physical mediums is also new, it’s fields all the way down :7

Sure, and if you want to study physics, at some point you need to experiment in the real world. If you want to study consciousness, that means you need to study the only medium we know that can evoke that phenomenon.

Again. We don't know it can evoke consciousness. It's an assumption of materialisam. If consciousness is fundamental you don't need that argument.

What the hell you mean by neurons holding consciousness lol.

This is known as "the neural correlates of consciousness": https://en.wikipedia.org/wiki/Neural_correlates_of_conscious...

By and large, the current scientific assumption is that everything required for consciousness exists within the physics of neurons, rather than, say, epithilial cells on your feet, or magic brainions that haven't been detected yet. Of course this is greatly simplified- a disembodied brain with no body probably can't sustain consciousness, but As far as we can tell, consciousness is an emergent property of brains which is wholly explicable using scientific methods.

> Marvin Minsky, Tegmark and Putnam have nothing to do with neuroscience, and have no authority to speak about the nature of consciousness.

Oh come on. Computer scientists and physicists are the pinnacles of humanity, who can speak with authority on absolutely everything, and have status that trumps every other kind of expert.

That's probably mostly fair, but then would you also agree that a hand-wavy piece of bloviation about purported quantum effects in a ubiquitous cytoskeleton component really doesn't have anything to contribute to the matter either?

Firstly, I would argue that it is very much in philosophy's remit until we can agree on a definition of "consciousness".

Secondly, if a misapplication of Godel's Theorem is used as evidence for the legitimacy of a hypothesis about consciousness, then it is perfectly valid for a philosopher to point out that misapplication.

I think we're looking at it from the wrong angle. It's really about psychology and theology.

There is no rational reason to even raise the question of why we have free will or consciousness in the first place. We do so because it appeals to our innate human desires that we evolved to follow, such as the desire to feel in control or have purpose.

These instincts are so strong and have such an ability to create biases in our mental reasoning that they cause otherwise very intelligent researchers to speak nonsense.

The more interesting question IMO is how and why exactly, from a neuroscience point of view, people consistently fail in their application of logic when it comes to questions about the meaning of life like this.

I think I get what you're saying. It's as if there was still a section of physicists clinging onto the idea of an ether, saying that it's critical to our understanding of electromagnetism and that the transmission of light cannot be explained without it. Meanwhile everyone else is saying we've moved on; the world makes sense without this.

Marvin's work in neuroscience, in his own words: https://web.mit.edu/dxh/www/marvin/web.media.mit.edu/~minsky... He invented a very successful sub-area of microscopy specifically to probe brains.

I care about Minksy's opinions in a wide range of science since he demonstrated himself as a capable general intelligence.

The ability to invalidate or critique a solution does not require knowing any part of the solution.

I recently explained my personal beliefs around how you square free-will and determinism (and subsequently consciousness) to GPT-4 and it told me this was the more formal name for it.

I posited that if you can observe and reconstruct the entire state of a complex system then you can predict future states- score one for determinism and no free will. But, we know there exists places that we cannot directly observe or perceive, aka quantum uncertainty, represented by σxσp ≥ ℏ/2 [1].

So based completely in theory, I figure the only way we square FW & determinism, is that free will exists somewhere/in a form we cannot directly observe, and it manifests as tiny influences that add up, in the complex system that is a brain.

This is the way more speculative part and it's more fun than anything to think about- it doesn't change the way I live my life buuuut-

Folded brains dramatically increase the influence a given region in space-time can have, simply due to the increased number of neurons. So our brains double as an antenna for some unseen influence that manifests through quantum uncertainty.

So when I explained this to ChatGPT it told me that OORT was very similar to this, but even the mechanism they use for it seems to be a stretch for me.

edit: But I do think that in order for neural networks to become anything other than outwardly really really good approximations of human minds, there needs to be a way to introduce a small amount of genuine randomness into their calculations, without utterly breaking them. I could see early attempts at doing this causing a form of LLM schizophrenia because the neural network wasn't resilient enough to the induced error.

[1] the standard deviation of position σx and the standard deviation of momentum σp is greater than or equal to half the reduced planck's constant

We don't actually know if quantum physics has real randomness or not. Quantum collapse is an unsolved problem.

> I could see early attempts at [introducing randomness] causing a form of LLM schizophrenia because the neural network wasn't resilient enough to the induced error.

1. This is actually already done. Temperature parameter controls amount of randomness.

2. Neural networks are quite noise resistant.

The temperature parameter doesn’t introduce any noise into the network evaluation.

Typically, what happens is that the network outputs a set of possible tokens with different probabilities, and a sampler picks from the top possibilities. Temperature determines how spiky its pick is; at zero it’ll always pick the top option.

You can get to this conclusion more directly by noting that computational complexity of any Turing simulator of anything more than a trivial system increases very fast as the precision of the initial conditions for the simulation increases. Even the shift map exhibits this phenomenon.

This can be an even more severe boundary for prediction than the actual measurement accuracy.

In the discussion about determinism vs free will, this leaves us with the situation that we can predict what somebody will do even if we assume perfect measurements, but will only be able to produce this prediction after the fact except for very short term predictions.

Stephen Wolfram calls this computational irreducibility.

Why would “my decisions are determined by sub-nuclear divine dice rolls” be any closer to free will than “my decisions are determined by algorithms operating on my sensory inputs and memories”? What’s more “free” about introducing that factor?

What does it matter why you can't predict the future state of a brain?

If you can't, we have free will. If we can, we don't have free will.

I didn't say "whether", I said "why".

At the moment, you can't predict the future state of my brain for more than one reason, one of which is that you don't have much information about the current state (precise information anyway, you may have an opinion about the average state).

>At the moment, you can't predict the future state of my brain for more than one reason, one of which is that you don't have much information about the current state

Do we not literally predicate our friendships and relationships on being able to predict the future states of minds? How long do you stay friends with the person who randomly shows up or doesn't, to any event you invite them to? Or whose tastes vary unpredictably from day to day, giving you no framework to contextualize them?

It's always very entertaining to nitpick a statement that has a caveat by quoting it without the caveat.

(No it isn't)

If why means because there is a real randomness: we have free will. If its just because of current complexlity, we don't have free will.

It also implies that we might life this life over and over forever.

This is a very common error people make when considering "free will". They mix in "predictability" to the concept. But predictability is not "free will".

If I give you a choice between a million dollars or a painful lingering tortuous death, you will with for-the-sake-of-argument 100% choose the million dollars, of your own free will. It is no less what you will for the fact that anybody can predict it; it is certainly what you will. Will you deny that is what you will?

Predictability also brings in a lot of contingency that people do not generally realize they are bringing in. If the universe is entirely material and there is no external reality, then good news! Your actions are already unpredictable. No conceivable machine built within the real physical universe could possibly fully predict your actions; you can prove this with some information theory considerations (the amount of information your actions leak about your internal state is not sufficient to nail down that internal state fully). So you have free will! Yet... if the universe is entirely material and there is no external reality, the universe may still be fully deterministic. Contrary to somewhat popular opinion, quantum mechanics is not intrinsically nondeterministic. It means you can't determine the outcome of certain events with any process we know from the inside, but the entire universe can absolutely have some sort of PRNG or something to determine everything that is going on and it could all be deterministic in ways that still work for QM. In which case, oops, no free will for you. So by this definition, the question is unanswerable from the inside.

Unpredictability is not free will either. If by some amazing, but physically possible, set of circumstances, the decision about whether to turn left or right came down to one 50/50 outcome decided by a quantum waveform collapse, that still doesn't give you "free will" about the outcome. You don't get to pick the outcome. It was undecided and unpredictable, but it wasn't decided by your "will" either.

If you're still not having enough fun yet, suppose "quantum" does "solve" free will. Which quantum outcomes make the difference? Suppose I build a perfectly-feasible quantum device[1] to flip a random coin, quantumly. Compare to a supposed quantum decision made "in" my "brain". How exactly is it that the latter is my "quantum free will" whereas the former is just a random decision made out in the universe?

Just labeling a process "quantum" doesn't do anything. It's just wordplay in the end, substituting one undefinable term for another and calling it progress. There's still a crapton of work to show that the "quantum" provides the mechanism for "something else" to meaningfully interact with the world[2]. My "will" is not "randomness". And boy-oh-boy is that "something else" a can of worms of its own.

[1]: https://www.youtube.com/watch?v=gwIGnATzBTg

[2]: https://news.ycombinator.com/item?id=41079700

I have to think through this take tbh :)

I liked the explanation of kurzgesagt around this topic but it feels weird if our unverse is not random.

Like moving the goal post to 'what would a real random source look like' 'would that source be called god'.

We have free will in either case. Whether or not our choices can be predicated is irrelevant.

are we really citing ChatGPT in comments now

There is a damn army of people doing this and I have no idea what they think they're contributing.

My personal conspiracy theory is it's ground work to set conditions for disinformation campaigns: the "I used an LLM/I used ChatGPT" people are there to make you look less critically at the other comments by giving a small queue that since they don't include those terms they just be more genuine.

> I have no idea what they think they're contributing.

I assume they are just young and see no harm in sharing something they thought was interesting.

This fad will die out eventually since it's redundant and provides no real value.

Sorry, I must be missing something, what's the problem here? I don't see OP citing ChatGPT, just that they were explaining their own belief system to GPT-4 and it responded by "simplifying" OP's beliefs into "orchestrated objective reduction". This is exactly the type of usage I would expect from an LLM; OP didn't use it to inform their decision, but to further examine the belief from another perspective or broaden their questioning around it.

> But, we know there exists places that we cannot directly observe or perceive, aka quantum uncertainty, represented by σxσp ≥ ℏ/2 [1].

> So based completely in theory, I figure the only way we square FW & determinism, is that free will exists somewhere/in a form we cannot directly observe, and it manifests as tiny influences that add up, in the complex system that is a brain.

These two things not only don't follow from each other, the first one actually all but refutes the second.

First of all, Heisenberg uncertainty affects all physical systems, but clearly not all physical systems are conscious.

Second of all, there is no pattern allowed to exist below Heisenberg uncertainty. That is, if you could determine exactly the momentum of a particle, the particle could literally be anywhere in the universe, with equal probability: there is no bias, it wouldn't be more likely to be here or there. So this is pure randomness, there is no "consciousness signal" you could extract from it.

Or, to put it another way, if our consciousness was a result of Heisenberg uncertainty, that would mean it's a purely random phenomenon, and every human at every time would be exactly as likely to type the next word in this comment, start running in a random direction, gouge out one eye, or any other thing they are capable of doing. There is, in a very fundamental sense, no way to get patterns or intention out of Heisenberg uncertainty.

Besides, the best way to square "free will" with determinism is Compatibilism. Every human is an automaton whose behavior is fully determined by genetic and epigenetic make-up and by everything they've ever learned and otherwise experienced. In a fundamental sense, my whole life's course was determined the moment I was conceived; but still, in any given situation, what I will do is different from someone else might do, because they have a different history and thus different values and biases. There is no magic that allows some "fundamental me" to "choose" how some electro-chemical processes will fire in my brain, any more than I could "choose" to emit electrons from the tips of my fingers. But that doesn't mean that I (the adult I am today) would do the same things Hitler did if I were somehow catapulted into his shoes today.

You've outlined what I reckon is the appeal of "quantum consciousness". I personally feel that randomness doesn't help to explain free will any more than determinism. I'm more inclined to believe that free will (in the strictest sense) is an illusion.

The problem with this approach is that even if you say that our thinking is non-deterministic because of true random effects on the quantum level, you still have to explain how deterministic calculations on random values make for free will.

You still have no influence on it, even if there is randomness involved.

You also have to explain why will is changed when the brain is damaged.

Really hard to justify free will (IMO) when a person's entire personality can be fundamentally altered by a bash to the head. What does "free will" mean if everything that makes you you can be changed with, say, a lobotomy.

It is, at best, an illusion and nothing more.

There is no illusion, and brain damage has no bearing on free will.

Free will is simply you making a choice, that's it.

If you want to argue about what 'you' means, feel free, but it doesn't really change anything here.

What is "choice"? Is it simply executing one of a set of possibilities? If you take such a general definition of free will, then a slot machine is manifesting it's free will to deny you a payout.

Making a decision either on impulse, intuition, or rational preferences.

> Is it simply executing one of a set of possibilities? If you take such a general definition of free will, then a slot machine is manifesting it's free will to deny you a payout.

Sure, or even just a dice. Except that's a rather silly definition of free will since it omits the 'will' part, i.e. thought.

> free will (in the strictest sense)

In what sense? Can you produce a strict definition, what is "free will", what is "illusion"?

This is a battle of definitions. Pick the definitions you like, and you can prove what you set out to prove.

The "strictest sense" is something like:

1. We have control over our decisions

2. Our decisions are independent of past events

I agree that this is pretty hand-wavey and open to semantics. But I don't think that there is any realistic, coherent way to interpret and reconcile the above two statements [edit - without resorting to some kind of non-physicalism e.g. God, spirit planes... ].

Not even a six-year-old would believe #2. It’s endlessly fascinating that there are people who do, but most people realise their past affects their future decisions.

> We have control over our decisions

That begs the question - what is "me"?

If I take "me" as the configuration of atoms in my brain, or simply the information if you will, then "me" is determining my future actions, therefore "me" is in control of my decisions.

Alternatively, I could define "me" as the whole system - the configuration (electrical signals), the hardware (brain, neurons), the physics. I think most "free will deniers" will say that physics is not part of "me", but I disagree - physics is not separable from matter and information, physical laws permeate everything, they are necessarily part of "me". You don't need any God here, this definition is as physical as it gets.

> Our decisions are independent of past events

Doesn't this require essentially random behavior? Sounds somewhat absurd ...

> I could define "me" as the whole system - the configuration (electrical signals), the hardware (brain, neurons), the physics.

I think that's fine. But imagine an outside observer who is privy to the current state of the system plus future environmental inputs. In principle, that observer is able to calculate the system's evolution exactly, and therefore predict all future decisions of the person-system. For many people this is contrary to the idea of free will. For how can something be "free" if it is bound by the laws of physics and is known in advance to any sufficiently sophisticated observer.

> In principle, that observer is able to calculate the system's evolution exactly

Which is fine, actually even more, it's confirming the presence of free will.

I have a Snickers bar in front of me. I like Snickers, but I also want my teeth healthy, keep diabetes away from me and keep my weight in check. I will decide to either eat it or not, depending on the current subjective trade-offs between my priorities. Acting on my wishes and priorities like this seems exactly in line with free will.

In your understanding, what behavior does an observer need to see to detect free will for you? The decision cannot depend on my goals of feeling good eating Snickers / keeping my weight in check? Based on what should free will make decisions? If it's on nothing, should the behavior with free will be random? That seems very counter-intuitive.

> if it is bound by the laws of physics

As mentioned earlier, this is no external constraint in my physicalist understanding of the world - laws of physics are an inseparable part of "me" as a system.

> The decision cannot depend on my goals of feeling good eating Snickers / keeping my weight in check?

You have conceded that your decision is pre-determined. Most people consider pre-determined events not to be "free". What are they even "free" from?

> You have conceded that your decision is pre-determined

Yeah, "pre-determined" by "me" (physicalist definition above).

Isn't the ability to determine your decisions exactly free will?

In my understanding of how the phrase is used, "free will" implies being free from pre-determination.

That's completely opposite to my understanding - free will means to be able to determine your own decisions.

I think I kind of get where you're coming from. Would you perhaps agree that the "free from" in your free will is "free from external influence"?

If so, consider the following thought experiment. Let's say I get to pick either an apple or an orange to eat. I'm confined to a white room and the state of the "me" system is such that I am pre-determined to pick the apple. Alice is standing outside the room in front of a button, which blasts an EM wave at my brain after which I am pre-determined to pick the orange.

So, my choice is determined by whether or not Alice presses the button.

Do I have free will when I make this choice?

What if, instead of Alice pressing the button, the wave is blasted with 50% probability?

> I think I kind of get where you're coming from. Would you perhaps agree that the "free from" in your free will is "free from external influence"?

I agree only to a degree. I don't consider free will to be a binary thing, in some cases you exert more free will, in other cases you are more strongly coerced to do something. Like when I talk to my wife during breakfast, she (external entity) is exerting influence on me, but I don't think it cancels my free will.

I don't fully understand the mechanics of your experiment, but assume that this EM wave just completely overwhelms the signalling in my brain - in that case, it doesn't seem like I have (a lot of) free will. But you can design less extreme thought experiments, e.g. where Alice causes discomfort to me which then influences me to favor one of the options more strongly than otherwise, where I still keep some decision making / free will.

I'd say that how much free will you exert is determined by how much your decisions are driven by your internal "me" state vs. the external influence.

> I'd say that how much free will you exert is determined by how much your decisions are driven by your internal "me" state vs. the external influence.

I reckon this definition would be very difficult to formalise meaningfully. Going back to the thought experiment in the white room, it might be that my choice is strongly influenced by whether the walls are cream or vanilla. My choice might be influenced by whether a certain smell or noise happens to occur just before I decide. I have a constant stream of inputs flooding into my system, affecting my choice-making. Some of those inputs involve deliberate actions by other agents (e.g. Alice pressing a button) whilst others are simply the goings-on of the environment. I suspect it is impossible in the general case to conclude that a choice is driven more by internal state than external influence or vice versa.

#2 is obviously false - and #1 is obviously true, to me. Whether or not you wrote this comment was a decision you made. The only way "free will" is even an open question is if you can't decide what "you" are.

If you only allow yourself to identify with the highest-level, most rational aspects of the decision engine you live inside, then that's a mistake which will haunt you with questions like "am I really in control?" forever. If you identify with a broader sense of your self, it's pretty obvious that you are making decisions, for both rational and irrational reasons. Your conscious experience is part of what it feels like for a human to make decisions.

Randomness just introduces branch points into the linear flow of deterministic states. Since you do not control the branch points or create them, this does not give you free will.

Randomness does not give you free will, any more than determinism does.

What do you mean by free will?

Exactly. If determinism is incompatible with somebody's personal meaning of free will, quantum dice rolls are hardly a solution. What they really need is to either find a religion or just shrug off philosophy and get on with their life, behaving as if they have free will even if they can't rationally justify it.

Quantum dice roll is _the free will_ in this context. So your free will is what sets the dice; as it is extraphysical, it will look like randomness in the physical world.

That's not quantum physics, that's just some sort of new-age religion. A new variation on the "brains are antenna for the soul" idea.

This is possibly one way to solve it, but I think there is a simpler way, following causal chains and the laws of thermodynamics.

We clearly have systems that can absorb energy for later use - creating a natural "pause" in the causal chain. Each of these pauses create a possible future that is not yet realized. The longer this energy is held, the larger this possibility space becomes.

Free will becomes that ability to hold the pause with intention, and then select from the different possible futures that have been created.

Determinism does not interfere with this in any way. The causal chains all follow the basic deterministic laws of physics. There is space for choice created by holding energy instead of immediately dissipating it.

No quantum mechanics required at all.

> free will exists somewhere/in a form we cannot directly observe

John Conway has a rather neat explanation of this in the Strong Free Will Theorem.

https://www.ams.org/notices/200902/rtx090200226p.pdf

Being neat doesn't necessarily mean it's correct, but it's compatible with what we know about physical reality, and solves some otherwise rather tough and paradoxical facts about experienced reality, so I'm a fan.

> how you square free-will and determinism (and subsequently consciousness)

I've never seen this as an issue. Even if something is fated, it's still you making that choice.

You ate whatever you ate for lunch yesterday. It's already happened. You still made the choice.

If it wasn't too old to be the case I'd think that article was just A.I. Slop or charitably something like technobabble from the Sternbach and Okuda era of Star Trek. "I can do math because I'm a thetan" shows that emotionally true stories can beat out factually true stories in science as well as politics.

Reply on news.ycombinator.com