Hacker News Reader

Comment by izzydata | Hacker News Reader

From my time reading and watching science fiction I've come to the belief that sentience is not binary. It's not a toggle that is either true or false. It seems like a fuzzy gradient with no definite threshold for becoming sentient. Maybe it is just from increased complexity or maybe there really are some key ingredients. I hope we figure this out in my lifetime though because it is a fascinating question.

View on HN

Totally agree, and its gradually growing from a animalistic fixated on the now state to full self-awareness. It can temporarliy collapse when fight and flight reflexes override the expensive reflection. It can permanently be lowered if a mental sickness (or a neuro-diverse adaption to more warlike states of humanity) gets triggered by stressfull circumstances.

And there is also a zoo of complexity when it comes to "sentience", meaning there are those who are "great thinkers in the subconcious" and can not explain there thought process. There are levels of sentience and almost automated task, were the subconcious kicks in and the sentient human just "dissolves" for a moment, like driving home in a car. Its a fascinating topic.

https://en.wikipedia.org/wiki/Blindsight_(Watts_novel) Obligatory Watts Reference, cause his book really does have a interesting look on the zoo of humanity.

The answer always seemed obvious to me. There is no threshold that makes us greater than the sum of our parts, no inflection point at which we become fully alive. We can't define consciousness because consciousness does not exist. Humans fancy that there's something special about the way we perceive the world, and yet we live in loops as tight and as closed as the hosts do, seldom questioning our choices, content, for the most part, to be told what to do next --Robert Ford, Westworld

People hate this take, but to me it is obvious. Our conscious experience has to ultimately be illusory, because we know that at least parts of it are verifiably false - like our perception of free will when really what actions we take are close to deterministic.

So consciousness is an illusion. Great. Then what are you experiencing right now? Who is experiencing it?

If what defines an agent who experiences something is something that takes in inputs, makes a "choice" independent of the base, deterministic interactions of atoms in their brain, and then executes on that choice - then nobody is experiencing it.

I agree that this is a difficult question - but it is equally obvious that the way we experience the world is not real. We feel that we have free will but we know we don't.

So it was preordained since the beginning of time that you would write this comment and that I would reply to you?

Quantum effects aside, yes.

To the extension that "preordained" has any meaning at all in a chaotic system.

This seems to be self defeating.

Since everyone’s words, including yours, would be those of the primordial demiurge (?) who got the ball rolling.

If ‘quantum effects’ are large enough to affect that causal chain then they would be enough to affect the course of human action.

Self-defeating? In what way?

> If ‘quantum effects’ are large enough to affect that causal chain then they would be enough to affect the course of human action.

Yes, but they are just manifestations of a deterministically evolving wavefunction.

So you assert all words, including yours, ultimately derive from a ‘deterministically evolving wavefunction’?

Assert? No, this is what all evidence indicates.

You certainly seem to be asserting something. Whatever any evidence indicates is only probabilistic and thus needs to be hedged with modifiers to be taken seriously.

And regardless, your words, and everyone else's, were still determined at the beginning of time, if your argument is true.

That's why this line of reasoning is self-defeating, since if one accepts it they will ignore you since they literally can have no effect on you, nor you on them.

How do we know this?

This statement is the hard problem in a nutshell.

The experiences are having themselves.

I don't understand the whole "free will is an illusion" thing. To me it seems obvious; I clearly do have free will because I choose to do things on my own volition. And the "me" who does the choosing is made up of largely deterministic electrochemical processes which respond to stimuli, with a bunch of feedback loops and other complexity. That is all my consciousness is, after all. I don't even know what it would mean for free will to be an "illusion".

The idea is that if you’re a bag of chemicals then any decision you make is entirely predetermined based on your current state and the stimulus you receive (potentially modulo some quantum randomness and plenty of chaos). If the decision is totally predictable and predetermined then it stops looking like free will, from a certain point of view. But it’s really just a question of the semantics of “free will”.

Trying to work out whether you could choose differently to your predetermined choice just isn’t a meaningful concept. Imagine you have a deterministic program which returns true or false based on some inputs. If it is modified to return the opposite of what it previously returned, it is no less deterministic.

I think it's clearly true that we have internal processes which cause us to take actions.

If those processes are entirely deterministic then that means every action you take is also deterministic - as in, you could not have acted otherwise.

When there are two choices laid out before you, you "choose" one in the same sense that an NPC in a video game chooses its actions: by running a complex set of internal logic.

"Free will" is the notion that you have the freedom to exert your will to choose courses of action. Even though you chose to do A, you could have instead have chosen to do B.

It sounds to me like you just explained the "illusion". You're made up of deterministic processes, but it feels like you're making a choice.

But I am making a choice? I am those deterministic processes. Whatever I decide is my choice. I feel that I am making a choice because I am making a choice, because those deterministic processes which make the choice are all that I am.

are you actually making a choice, or is it just an output of those deterministic processes? The choice was deterministic.

The output of those deterministic processes _is_ the choice though? I don't understand how there's any semantically sensible way in which I, as a collection of largely or entirely deterministic processes, can process input and my internal state and reach a choice, where that choice isn't "actually" a choice or isn't "actually" made by me.

There was a good podcast on this recently if you want a deep dive into the philosophy behind the statement.

To give you a jumping off point, 'illusion' has a specific definition in the non-dualist domain so you want to make sure you aren't bringing your own from home. You also smuggle in an 'I' and a 'me' which should be breakpoint-inducing words to mention in a philosophical conversation. The people claiming free will is an illusion have rationally cancelled 'I' out of their statements.

You've basically made the argument against free will in your comment but your definitions are loose so you parse it as an argument for free will.

https://www.samharris.org/podcasts/waking-up-conversations/d...

If you accept that a brain is nothing more than a blob of electrical/chemical circuits that respond to stimuli, then the brain is entirely deterministic.

You think you're deciding things, hence you think you have free will. But all those decisions are based on deterministic circuits, therefore you're not actually in control. Your ability to decide things and act in free will is an illusion.

But that's the thing, right? Those deterministic processes _are_ me. There is nothing more to me. So I think I'm deciding things because... I am. I am the only thing in control, because the only thing I am is a set of deterministic processes. If my deterministic circuits are making choices, _I_ am making choices.

Then that means you don't have free will. Consciously, you might feel like you do, but it's all deterministic.

But "free will" and "deterministic" just don't seem at odds. It's not just that I "feel" I have free will. It's that I make choices freely, those electrochemical processes are _all there is_. Whenever those electrochemical processes reach a decision, that decision is purely a product of my free will, because those processes are me and I am those processes.

Wind the clock back to a time when you made a decision. Is it possible for you to make a different decision if you had the exact same mood/context/environment/knowledge and all the sub-atomic particles were exactly identical with the same spin, etc? Current dominant philosophical thought points towards a big NO.

Well current scientific thought says yes because measuring the spin in a particle in a superposition is truly non-deterministic from the perspective of observers, but I don't think that's relevant and I'm happy to think about the brain as an entirely or mostly deterministic thing, so I don't think this is too relevant, but we're not in a clockwork universe.

>Well current scientific thought says yes

Link to evidence?

Well, quantum mechanics describes the universe as a wave function which describes a probability distribution. Most physicists who work on quantum mechanics seem to believe that wave function collapse is a truly nondeterministic process; the only real alternative, from what I understand, is hidden variable theories, which we know imply nonlocality thanks to Bell's theorem.

There's the caveat of Everettian quantum mechanics, where the universe is described exclusively through the wave function which evolves completely deterministically. However, this still causes the observer to experience a measurement outcome as truly nondeterministic; there's just one version of the observer for each measurement outcome.

Here's the Wikipedia article on quantum indeterminacy, which I'm sure you can use as a jumping-off point for further reading: https://en.m.wikipedia.org/wiki/Quantum_indeterminacy

EDIT: Though while this topic is fascinating, I don't think it's extremely relevant to my own thoughts on free will or whatever. I'm perfectly happy to discuss a hypothetical brain which is truly deterministic, and I don't think I would consider a consciousness simulated on a deterministic computer to have any less "free will". The consciousness in the computer would _be_ the circuitry and program. Of course it couldn't make different choices twice given the exact same state and input; it wanted to make the choice it did because of its state and inputs. I don't understand how the choice is any less free just because it's deterministic.

The quantum substrate being non-deterministic doesn't mean that free-will exists/doesn't exist. It has to be shown as such.

What is happening at the quantum level, to the extent that we can understand it, isn't directly translatable into a non-deterministic world at the higher macro level. And so, we do expect and observe a certain level of determinism in our macro scale world. For e.g. A topic relevant to this community - Computing. For the vast vast majority of cases when thousands of people execute the same program on the same or similar hardware at different times in different regions we do expect, and observe the same deterministic result. The outliers to this are primarily due to damage to the processor/equipment, software bugs, or other known factors (including alpha particles/radiation, etc).

Ah, I see what you're saying. I interpreted what you said to mean that current philosophical thinking is that the universe is deterministic, but you're saying that even though the universe is nondeterministic, the processes in the brain are completely deterministic. This feels like a purely empirical question rather than a philosophical one; I don't know whether there are processes in the brain which depend on quantum measurement outcomes. Though I would be kind of surprised if there's nothing at all; something as simple as putting a polarization filter in front of your eyes would make which exact photons are hitting your retina determined by the outcomes of quantum measurements, which seems to mean you'd be introducing truly nondeterministic noise into the system. But I'm also completely open to the idea that in general, quantum measurements don't affect the processing of the brain or sensory inputs in any way which affects decision-making, if that's what empirical investigation of the matter shows. Whether the brain is deterministic or nondeterministic doesn't really have any impact on the way I think about the concept of free will.

I think Everettian interpretations are secretly quite popular among physicists because the Copenhagen interpretation doesn't really get you that far. In Everettian QM, the evolution of the entire system remains deterministic.

This raises the question of "can we identify something that is more sentient than us?". I think we can spot things that are less, just look at an insect. But what would more look like?

Not merely more intelligent, that's probably not very difficult. More... alive or something.

Sometimes I wonder if people on LSD are more sentient.

No, they're sensing different than usual. It provides "new" experiences, not "more" experiences. You'll have more memories, but they aren't "more elevated" than the ones you're accustomed to - the brain doing the experiences is the same one, even if working differently.

It does dedicate more processing power for pattern matching, so if consciousness (however defined) does correspond to that on the physical layer ...

This is the correct way of seeing it, but lots of people seems to feel personally hurt when someone rebuts the claim "It's scientifically proven!" with their own data, failing to understand that that specific process is exactly what science is.

I agree. Sentient or not, I think the fact that we now need to start seriously considering these questions, and not just as some far away hypothetical exercise anymore, is quite impressive in itself

This seems to not be a very useful distinction (binary vs gradient) though, because in practice the question in the gradient scenario moves to “at what point along the gradient do we demarcate sentience”.

I think the idea is that the entire gradient is sentient to varying degrees. Two neurons interacting could be sentient, but is not enough to us as humans to have any meaningful interaction with. A cpu with two gates is way less useful than a cpu with a billion gates because one can perform more useful calculations to us more quickly.

A creature with a brain that has 10 times as many neurons as a human probably isn't 10 times as sentient, but it might enable than to perceive reality in a way that a human never could imagine.

I think the problem might be that we are wanting to define sentience in a too human centric way. Just because something like an insect can never perceive reality like we do as people doesn't necessarily mean it is not a simpler form of sentience. An insects brain probably doesn't dedicate any neurons to allow it to perceive suffering or itself, but it clearly perceives something.

I don't know though. Just rambling here.

Reply on news.ycombinator.com