Live data from Hacker News

Is artificial consciousness achievable? Lessons from the human brain

arxiv.org

591–600 of 604 posts

Re: Is artificial consciousness achievable? Lessons from the human brain

#591
post #576

Earlier quoted context omitted.

> I say they pump the intuition because it is not actually an established fact that the premise is correct, but to be clear, I think it is, at least for human consciousness. I don't see what's being pumped . As we know from the incompleteness theorem, some truths cannot be proven. The fact of our awareness is another that's often doubted by the same people that doubt the divide separating qualia and conceptual knowin…

You say you are concerned with the challenge of pulling our cultural head out of [what you perceive as] the sand of implausibility - well, in that case, I'm the sort of person whose concerns you should be addressing: I don't think the hard problem (if there is one in the form it is conceived of by its proponents) is any harder for materialism than it is for any of the alternatives, but I am willing to give all due co…

I neither recognize in my discourse a sermon targeted at the choir (our psalms are way past this), nor do I feel an obligation on my part to convince you. I don't even register as a blame my failure to communicate my "belief" to you, especially since you say that you were already acquainted with those arguments, which you've looked into in the past, but also failed then to integrate. I doubt that I'd fare any better than your past attempts. For that matter, I also doubt that you're my target audience. But that's totally fine.

I'm indeed interested to blow fresh wind to a more parsimonious direction to our cultural view of reality. Despite this, or perhaps because of this, I don't see much value in debating the subject to the point of sophistry. Philosophy is not my day job. My interest in it is as a tool that informs a practical orientation. Does matter give rise to consciousness? Mounting evidence points to the contrary. To me this is practical. If what you're after is ironclad proof, you won't find it and personally I don't see the practical point.

Opposing views to the knowledge argument take two general forms. Either they deny the distinction between phenomenal experience and conceptual knowing (e.g. Dennett). Or they concede that divide, but posit that phenomenal experience could still be physical in nature, but built into this is a requirement to account today, for some unknown property of matter, of which we have exactly zero evidence and that we might never discover (e.g. Alvin). I personally see either as a cop-out. You're free to assign them the value you see fit and even to think that they successfully reduce the idealist intuition to mere beliefs. I think I recognize glimpses of the tangent the discussion is taking and would rather avoid venturing in ever speculative terrains, that hinge on hopes that some day, some big reveal in physics will be retrofitted to what is currently a baseless, hasty, and problematic assumption. From experience, this tends to drag on and people interested in that sort of exercises tend to already have a somewhat significant, even if tacit, commitment to materialism. No argument will be sufficient, as new pseudo-counters are sought out to justify the hold out.

I think it's best for me to conclude this exchange here. I'll echo the O.P.'s suggested paths of exploration, books/articles/videos by Bernardo Kastrup and Donald Hoffman, who contrary to me, have invested copious amounts of energy to make the modern views of idealism accessible. Their arguments go into details about what I've only evoked as bullet points throughout the thread. They attack the subject from physics, biology, philosophy, neuroscience, and sometimes even venture into the spiritual (why not, if consciousness is indeed fundamental).

Thank you for an interesting discussion. I hope you find the answers you're looking for.

Re: Is artificial consciousness achievable? Lessons from the human brain

#592
post #228

Earlier quoted context omitted.

If we abuse a dog we have no way to restore it to its previous state. With computer programs we have a perfect time machine, so any thing that one may call "horrible" can be done and then undone without any moral consequences. This btw also is the answer to the question of evil in religion, god can do whatever he wants without being evil, because it is effectively all in his imagination, and for the people living in…

There are many leaps here, but the biggest leap is how reverting abuse later nullifies suffering incurred.

Because the sufferer is back exactly at the state before suffering, without even a recollection of suffering. And all the suffering was merely number multiplication.

Re: Is artificial consciousness achievable? Lessons from the human brain

#593
post #421

Earlier quoted context omitted.

Why not? I can replicate a book in a foreign without understanding the language?

without understanding the language, you still understand that it's a book, it has symbols which represent language, and those symbols can be replicated. we don't understand what consciousness is or how it is achieved, so discussing how to recreate it isn't really a conversation we can have - we can only discuss how to create a simulacrum.

That’s not because it’s some deep philosophical problem, but because people refuse to define what they mean by consciousness.

Do they mean memory? Attention? Awareness? Self awareness? The inner voice? Qualia? All of these can be explained somewhat, but they keep moving the goalpost

Re: Is artificial consciousness achievable? Lessons from the human brain

#594

Earlier quoted context omitted.

In many scenarios, the observer is a machine or tool, not a human mind. And of course there's that whole aspect of replication along with that "scientific method" thing. If science was simply the act of humans making observable predictions and telling them to others, then there would be no difference between "science" and "personal anecdote". I also don't understand why the mind is relevant. We are trying to prove so…

> that whole aspect of replication along with that "scientific method" thing So uh, religion has been replicated quite a lot. We have historical records of it. We've seen an unprecedented revolution from religion, including science. And we've seen our pinnacle of civilization beginning to collapse since most people abandoned God. How much more proof/evidence/anecdata you need? We still track time in years since Jesus…

I have not heard of this replication before so I would be glad to see some examples of this! I mean I'm fairly convinced Jesus did exist, I'm just not convinced that they had any of their spiritual powers.

I have definitely not seen our civilization start to collapse though. I'm not even sure what that would look like (maybe a transition to a low-trust society or something)?

Of course, I do not believe everything can be tested by science, but my belief that religion specifically can be tested is because religion describes the most powerful forces in the universe. And not only that, humans can interact with these forces! So we should be able to detect these forces by observing how humans behave when they interact with these forces.

Re: Is artificial consciousness achievable? Lessons from the human brain

#595

Earlier quoted context omitted.

In many scenarios, the observer is a machine or tool, not a human mind. And of course there's that whole aspect of replication along with that "scientific method" thing. If science was simply the act of humans making observable predictions and telling them to others, then there would be no difference between "science" and "personal anecdote". I also don't understand why the mind is relevant. We are trying to prove so…

> In many scenarios, the observer is a machine or tool, not a human mind. An unconscious, non-experiencing mechanism is not an observer in the way the term “empirical”[0] is meant—to observe is to experience. > I also don't understand why the mind is relevant. See above. (I do not think I really understood the rest of your comment.) [0] https://en.wikipedia.org/wiki/Empirical_evidence

You can observe the state of a machine that is expected to derive it's state from an event or state of another object. For example, a video camera derives it's state from the light rays entering the lenses.

I'm asking for some video evidence of religion. So really I am asking for an opportunity to observe a state of a machine, albeit a very specific state. I suppose you could argue this is just a very roundabout way to indirectly experience religion.

Re: Is artificial consciousness achievable? Lessons from the human brain

#596
post #421

Earlier quoted context omitted.

without understanding the language, you still understand that it's a book, it has symbols which represent language, and those symbols can be replicated. we don't understand what consciousness is or how it is achieved, so discussing how to recreate it isn't really a conversation we can have - we can only discuss how to create a simulacrum.

That’s not because it’s some deep philosophical problem, but because people refuse to define what they mean by consciousness. Do they mean memory? Attention? Awareness? Self awareness? The inner voice? Qualia? All of these can be explained somewhat, but they keep moving the goalpost

I would say it's not because people refuse to define, it's that science has not defined (and cannot define it)

Re: Is artificial consciousness achievable? Lessons from the human brain

#597
post #591

Earlier quoted context omitted.

You say you are concerned with the challenge of pulling our cultural head out of [what you perceive as] the sand of implausibility - well, in that case, I'm the sort of person whose concerns you should be addressing: I don't think the hard problem (if there is one in the form it is conceived of by its proponents) is any harder for materialism than it is for any of the alternatives, but I am willing to give all due co…

I neither recognize in my discourse a sermon targeted at the choir (our psalms are way past this), nor do I feel an obligation on my part to convince you. I don't even register as a blame my failure to communicate my "belief" to you , especially since you say that you were already acquainted with those arguments, which you've looked into in the past, but also failed then to integrate. I doubt that I'd fare any better…

Firstly, I have only just now noticed that you posted two consecutive responses a couple of days ago, and I only responded to the second. In the first, you write " conversely, if reality is taken to be fundamentally consciousness, then everything is consciousness... This position isn't subject to the hard problem since everything in reality is all just one thing, consciousness."[1] Well, we can also say that a materialist theory of consciousness would not be subject to the hard problem, because, in that case, there clearly isn't one, by definition! This is even before we get into the question of what, if anything, it means to say that reality is fundamentally consciousness.

Turning now to your latest post, in your third paragraph, you offer some sort of response to the knowledge argument issue, but it both misrepresents the full scope of materialist responses to the argument, and, more relevantly here, completely misses the point to which that argument is being used in this discussion.

While the latter renders the former moot, I will, for completeness, say something about it. Firstly, I know (from private correspondence) that Daniel Dennett considered "What RoboMary Knows" to contain the essentials of his response to the knowledge argument. In it, he argues that if we had a different neural architecture - one in which we could directly examine and modify the detailed physical state of our brains - then learning what it is like to see colors could be done discursively. The fact that we humans cannot do this is, therefore, a contingent fact of biology which poses no challenge to materialism.

Secondly, you are once again completely mistaken in your guesses about what I think. Personally, I don't feel that the opponents of materialism have shown that consciousness will prove to be inexplicable without new physics, any more than are other biological processes such as metabolism or reproduction. Dennett's response to the knowledge argument is not predicated on new physics, and (while I don't think they are very helpful) neither are the arguments from the phenomenal concepts wing. Part of the rhetorical genius of Jackson's argument is that it nudges readers down the path of thinking that materialism will need new physics to prevail, but, as shown above, no such conclusion is warranted.[2]

Thirdly, though I'm not positing any new physics, I can still note that your characterization of those views as postulating something "of which we have exactly zero evidence and that we might never discover" is rather breathtakingly ironic, given how you are going about justifying idealism. As for avoiding debating the subject to the point of sophistry, I think that would be very helpful here.

As I said, though, this is moot, as it misses the point. I had hoped to forestall this outcome by pointing out that the question I posed is not predicated on any assumption of the truth of materialism, but it seems I should have said more about why that matters, so I will do so now. The question is this: why does the knowledge argument, when cast in terms of complete knowledge of idealism, not establish that there is a hard problem for idealism, just as the corresponding physical-knowledge argument allegedly does for materialism? Instead of replying to that question, you have offered some sort of defense of the knowledge argument against materialism - but the more strongly you promote the latter, the more strongly you support the view that the corresponding knowledge argument against idealism needs a substantive response (I have, of course, just referenced an argument that it is not actually a problem for materialism, but if you were to seize on that argument for your own purpose, it would raise the question "what hard problem?" - if the best and arguably only argument for there being a hard problem is no more (or no less) applicable to idealism than it is to materialism, you cannot use it to establish that there is a hard problem for materialism alone.)

If idealism really does provide an explanation of consciousness, you should have no difficulty responding to this issue, but instead, we have circled around it three times now without getting any closer to a solution. As you yourself put it, the hard problem is the linchpin of idealism: without it, all your arguments for it being the only viable non-materialist option are beside the point. [3]

Well, so much for the third paragraph, but quite a bit of your latest reply is taken up with other, incidental, matters, such as whether at least some of your arguments amount to preaching to the choir. Let's look at a definition, and from Merriam Webster, we have "to speak for or against something to people who already agree with one's opinions." I think we can leave it to third parties to decide for themselves whether your claim that materialism is obviously false for those who "get" qualia (in the right way, of course) fits that definition. Furthermore, when we put together your statements that, on the one hand, that you are are attempting to pull our collective heads out of the sands of implausibility and blow fresh wind to a more parsimonious direction to our cultural view of reality, while on the other, that you don't have to take into account (or, apparently, respond substantively to) the apparently awkward questions I have been raising, then we can see that you are more interested in the one-way delivery of ideas than in dialogue, which comes across as rather preachy.

Your posts have been moving in the direction of a motte-and-bailey argument. In your first paragraph of your first post in this thread, you were squarely in the bailey, writing "perhaps for some it's indeed a matter of "feelings". But for others it's a conviction built from reasoning that leads to self-validating and irreducible truths [my emphasis]. If you seriously go into this, the only possibilities left once you've dug and eliminated all mistaken assumptions are not material", but now, with "does matter give rise to consciousness? Mounting evidence points to the contrary", you have at least one foot in the motte. I am not, as you put it, after ironclad proof, just arguments strong enough to justify the certainty with which you have, at least up to now, presented idealism.

I thank you for your kind wishes in your last paragraph and I wish the same for you. I imagine you will have more success in that than I will, as I am quite demanding in what I expect in an explanation, and the mind is a hard problem, even if it is not the hard problem.

[1] https://news.ycombinator.com/item?id=40479264

[2] Penrose has a different argument for that, one that is most commonly rejected on account of its assumption that materialism based on known physics entails that human minds must be logically consistent reasoners.

[3] At least one of those arguments - the one from parsimony - is problematic in its own right: the one and only essential property that any hypothesis of the mental needs in order to prevail is that it actually explains minds, and, so far, we have seen none from any position, materialism included (I am well aware that quite a few physicists think physics will continue to deliver parsimonious theories (The Elegant Universe, and so forth), but that, too, is a belief for which even the inductive form (so far, it has been that way) has a rather obvious confirmation bias problem.)

Re: Is artificial consciousness achievable? Lessons from the human brain

#598
post #587

Earlier quoted context omitted.

You say you are concerned with the challenge of pulling our cultural head out of [what you perceive as] the sand of implausibility - well, in that case, I'm the sort of person whose concerns you should be addressing: I don't think the hard problem (if there is one in the form it is conceived of by its proponents) is any harder for materialism than it is for any of the alternatives, but I am willing to give all due co…

> "you can't learn what it is like to see colors from any physics book." Why is it an argument against materialism? To know what it is like to see colors you need to put your brain in a state in which other people's brain gets when they see colors. Physics book simply does not do that by itself. But if you use learned physics, to electrically stimulate the right neurons in your brain, you can learn what it is like to…

The short answer is that, when the argument is fully set out, one of the conditions is that stimulating one's optic nerve does not count, but it often goes unmentioned. This is rarely an issue (and is easily corrected if it is), as it quickly becomes apparent that allowing stimulation does not even present a prima facie problem and so no-one on either side of the fence finds it at all interesting or useful - the anti-materialists are not interested in this question because they will agree with materialists that if stimulation is allowed, then Mary (the protagonist in the thought experiment) will experience colors, and the materialists are not interested in it because the anti-materialists are not using it.

Nevertheless, it is a good question as to why the knowledge argument would be seen as so persuasive by so many, including by quite a few materialists, who seem to me to go to unnecessary lengths to get around what is not even a problem (illusionism, for example, where it is claimed that phenomenal experience is merely an illusion - but then, as the anti-materialists ask, who is being fooled, and about what?)

Perhaps the first thing to say is that the argument is not presented as the stark observation that you cannot learn what seeing colors is like from reading a physics book; that's what you end up with when you whittle it down to its essentials.

Secondly, there are a couple of features of the argument which make it easy to let its difficulties slide right on by. One of these is what I have been going on about here: by framing the argument in terms of a knowledge of physics, it is easy to miss that it applies to any academic knowledge, not only physics or the physical sciences, and thus including any non-materialist theory. The second is that it uses the word 'knowledge' for two different forms of retained information: sensory information which is acquired directly from sensory experience, and linguistic information which is encoded in sensory information without being that sensory information itself. These are separate domains (the information content of a word is not the information content of the sound when it is heard or of its appearance when read), yet, by referring to both of them as knowledge, the argument invites the reader to accept the way it equivocates between the two, which it does when it points out that the knowledge that can be acquired linguistically does not include the 'raw feels' (yes, that is a term used in the philosophy of mind) of sensory information.

Having said all that, I am still surprised how many people think the argument shows there is a problem for materialism. After all, the physical sciences have explained many sorts of complex phenomena (hurricanes, for example), and no-one thinks that the act of explaining (or learning the explanation of) how hurricanes work should actually create one - yet anyone accepting the knowledge argument is tacitly accepting that if the phenomenon being explained is sensory experience, then an explanation should produce the phenomenon being explained! (but only if it is a materialist explanation!) I guess it somehow doesn't seem paradoxical to many people when both the phenomenon being explained and its explanation are in one's mind.

I think you would be very surprised by how many papers have been, and are still being written about this argument or something following directly from it. Because I have read some of them on academia.edu, I frequently get notices about more. Interestingly, Frank Jackson himself has changed his mind on the matter, but that has had no effect on how influential his argument still is (and, to be fair, it shouldn't.)

Re: Is artificial consciousness achievable? Lessons from the human brain

#599
post #590
post #587

Earlier quoted context omitted.

> "you can't learn what it is like to see colors from any physics book." Why is it an argument against materialism? To know what it is like to see colors you need to put your brain in a state in which other people's brain gets when they see colors. Physics book simply does not do that by itself. But if you use learned physics, to electrically stimulate the right neurons in your brain, you can learn what it is like to…

Even in normal experience the brain is being stimulated. So theoretically, you wouldn't need eyes to experience color. You would only need to replicate the physical properties at the onset of brain activity. Those are called neural correlates of consciousness . However, there's the byproduct as the result of that activity, the experience associated with it. If reality is fundamentally material, there are two possible…

It is true that neither 1 nor 2 have been successfully demonstrated, but there is quite a double standard here: you claim that idealism has the answers to questions about the mind, but you have not shown it explaining anything. Deducing the reality of idealism from the absence of answers to 1 or 2 is exactly like saying, in 1950, that as biology has not identified the biochemical correlates of cell reproduction, life must be fundamental (I know some people hold this view today, but it is at best a niche view that generates little controversy in either philosophy or science.)

Re: Is artificial consciousness achievable? Lessons from the human brain

#600
post #586

Earlier quoted context omitted.

You seem to insist on painting dualism/idealism/… as a contender to physics. This is a category error that can only be explained by implicitly treating physics as religion. It is not uncommon—many of us are from a generation that is freshly atheist after generations of religious adherence, so once you encounter physics it is tempting to use it as an outlet for all that bottled up religiosity—but is wrong. Physics is…

> If the aforementioned stress is the root cause of the physical consequences, then that is materialism-inconsistent. E.g. stress increases production of certain hormones, raises blood pressure etc., which over time can be harmful. How is this materialism-inconsistent? > Have you heard of the Chinese room thought experiment or the concept of philosophical zombies? Yes, i don't find them particularly convincing. Chine…

> stress increases production of certain hormones, raises blood pressure etc., which over time can be harmful. How is this materialism-inconsistent?

Because this roughly simplifies to “stress causes disease” with extra steps inbetween. Materialists are allergic to such claims.

> because no one argues that computer carrying out the arithmetic operations does the thinking. The program running on the computer does the thinking, and it does not matter what mechanism is used to implement the computation [1]

That’s a theory in philosophy of mind, too—a non-falsifiable speculation, like the rest of them.

> Creating software that does things similar to what mind does, is now not simply a science but already a field of engineering, so i don't understand what do you mean.

You said “science is concerned with an implementation of device that behaves like a human would”. I said what we are talking about here is not about that, but about whether that implies there is a consciousness or it’s an unthinking machine. Whether outputting things like a human is enough to consider software conscious, thinking and self-aware in a human-like manner (and thus we are abusing human-like thinking, conscious and self-aware creatures by using ML the way we do), whether consciousness is the substrate as opposed to material world (like Schrödinger, among others, believed), etc. That’s the point of what materialism/idealism/dualism is about, making computers and programs is irrelevant.

> If we manage to simulate brain with high enough accuracy and it does not produce a behavior similar to human behavior, that will be a proof that you are right and computationalism is wrong

No, it only means there is a program that produces behavior similar to human behavior—the “proof” you imagine is not a proof, which can trivially be demonstrated logically (as Chinese room shows). Manipulating syntactic tokens as an LLM does does not mean understanding and manipulating ideas like a human does, unless you hold a particular non-provable and non-falsifiable theory of mind.

Post reply on HN