Hacker News Reader

Comment by ppod | Hacker News Reader

Is it really surprising that we have a first person subjective experience? We know that we are incredibly complex things, constantly integrating and acting on very complicated external stimuli. Such a system should have references to its own body and its own neural states, its train of reasoning should frequently include itself, its focus will drift forward and backwards in time... this is just how a system like this would work. If the system communicates about its state then its language should have referents to these internal states, referents like "experience", and "feels like", and "I understand". Is that surprising? Wouldn't it be surprising if it wasn't like that? I don't think you need to invoke an essentially mysterious "conscious" property of the mind to explain that.

View on HN

People used to make this same argument about why things fall to the ground when you let go of them. "Is it really surprising? Why wouldn't they just fall down? What kind of answer do you want?". You have to allow yourself to be puzzled by things.

So yes, it is surprising. You don't seem to understand the most basic principles underlying the so-called hard problem. Nowhere in your description about "complex things" and "integrating" and "complicated external stimuli" and "communicating about its state" have you even attempted to point out why it actually DOES feel like something to be that system. Why does all this stuff actually lead to the feeling of pain, or intense grief, or pleasure, or anything at all. That is the fundamental mystery - why does it feel like anything at all when this "stuff" happens, and what is the nature of the feelings themselves. There are tons of mysteries here that you can't gloss over by saying "well why wouldn't it??".

I am very familiar with the literature on the hard problem. It's remarkable how often people respond to this line of argument by putting the words "feeling" or "understanding" in italics. I don't know what kind of causal explanation you want for pain or grief. They are processes in an organism, and we construct models of the self to help us get through the day, as Dennett has elucidated many teams.

To answer your first question: I believe in puzzles, not mysteries. When we don't know how something works (like a spray bottle), we call it a puzzle, unless it challenges some sacred theological or humanist tenets, in which case we call it a mystery. I don't agree with the comparison to gravity --- mechanics has well defined and refined concepts that allow us to create more general and parsimonious models with precision. "Consciousness" and "feeling" and "experience" are folk terms with neither a precise definition nor a clear process for arriving at a precise definition.

> To answer your first question: I believe in puzzles, not mysteries. When we don't know how something works (like a spray bottle), we call it a puzzle, unless it challenges some sacred theological or humanist tenets, in which case we call it a mystery.

Rather than the boundary between the biological and "non-biological" brain, I am interested in the notion or aspect of consciousness that relates to why people behave the way they do, think the way they think, and are not only largely oblivious to it (the idea of examining behavior and the ~motivations behind it) but commonly hostile to it, sometimes extremely so. Or even more interestingly, the ability to easily notice the behavior in others, but utterly incapable of seeing the same thing in oneself.

For someone that has no background in the subject, would Dennett be a good place to start, or could you suggest any other names?

> "Consciousness" and "feeling" and "experience" are folk terms with neither a precise definition nor a clear process for arriving at a precise definition.

I agree, but I "feel" like this is where progress can be made (unless, it already has and I'm simply not aware of it).

An attempt to give a hint at the aspects that I'm thinking of: reality, at least as far as humans perceive it, consists of observations of physical events, or at least we know that much for sure. One thing that the layman overlooks is that there are far more observable dimensions involved in reality than are commonly discussed, some of which I suspect are easily observable to some people, but practically invisible to others (a terrible analogy might be how humans can see one segment of the light spectrum, while other mammals can see others, and machines can see more). Similar to the formerly "invisible" portions of the light spectrum, are there other "dimensions" (for lack of not knowing a more proper term) that we can't currently observe, but with the proper experiments, consistently and with decent accuracy "detect", which might lead us in more specific directions when studying the mind? For example, if we can identify a highly reproducible but currently completely unexplained phenomenon (behavior, reaction, etc) common across large quantities of people (but perhaps not among other groups), might ML algorithms run on brain scans of subjects under these conditions find anything interesting? (Possibly relevant: Moran Cerf: "Decoding Thoughts and Dreams Using In-Brain Electrodes" | Talks at Google https://www.youtube.com/watch?v=EVj3sU37gdI)

I also realize this may very well be a very stupid idea, so anything substantiating that possibility is also welcome.

EDIT: Or, another different way of thinking about it is, an attempt to form an aggregate (across all people & cultures) definition/enumeration of self-awareness.

> I am interested in the notion or aspect of consciousness that relates to why people behave the way they do, think the way they think, and are not only largely oblivious to it (the idea of examining behavior and the ~motivations behind it) but commonly hostile to it, sometimes extremely so. Or even more interestingly, the ability to easily notice the behavior in others, but utterly incapable of seeing the same thing in oneself.

I completely agree that this is maybe the most interesting general field of inquiry out there, I just think that the consciousness literature/debate has almost nothing of interest to say on it. I think the fields of interest are evolutionary psychology, neurobiology, and at a higher level the interface between sociology and behavioural psychology.

>a terrible analogy might be how humans can see one segment of the light spectrum, while other mammals can see others, and machines can see more

I like analogies like this and am totally on board with you here. But again, things like 'Mary's Room' are not at all interesting or detailed addresses to this when we know so much about how brains and bodies and artificial neural networks can hold and use so many kinds of knowledge at different levels.

You're getting kind of towards Penrose territory at the end there with suggesting that there could be senses or phenomenologies that we currently know nothing of the mechanisms of. I agree that this is true in principle but I think that the evidence that we have from biology and from artificial intelligence suggests that we could probably explain all of our abilities just with physical properties that we already understand. I wouldn't rule out some quantum component though. I think that it's likely the problem is one of scale and complexity rather than a qualitatively new kind of process. The brain is not only huge but is designed by a messy ad-hoc evolutionary process that is hostile to interpretation by our own reflexive symbolic investigation, and I'd say that's where the difficulty lies. I completely agree with your line of thinking though.

> I like analogies like this and am totally on board with you here. But again, things like 'Mary's Room' are not at all interesting or detailed addresses to this when we know so much about how brains and bodies and artificial neural networks can hold and use so many kinds of knowledge at different levels.

Mary's Room demonstrates that's there's more to knowing the world than just third person descriptions. There's also experience.

Whether you want to call experience a form of knowledge, or something else like an ability is the crux of whether that particular argument works or not, since it's based on whether knowing all the physical facts leaves some knowledge out (experiential).

But I think it works to show that our physical theories of the world can't capture our experience of the world, because one is abstract and mathematical, and the other is how we perceive and feel. So nothing about biology, physics or ANNs bridges that gap, I don't think, since they're expressed in third person terminology. That's the hard problem.

> I completely agree that this is maybe the most interesting general field of inquiry out there, I just think that the consciousness literature/debate has almost nothing of interest to say on it. I think the fields of interest are evolutionary psychology, neurobiology, and at a higher level the interface between sociology and behavioural psychology.

Based on what little I know, I would agree, but lots of obviously knowledgeable people in the thread so thought I'd throw a line out there.

> Mary's room

https://en.wikipedia.org/wiki/Knowledge_argument

"Roughly thirty years later, Feigl expresses a similar notion. He concerns himself with a Martian, studying human behavior, but lacking human sentiments. Feigl says: "...the Martian would be lacking completely in the sort of imagery and empathy which depends on familiarity (direct acquaintance) with the kinds of qualia to be imaged or empathized" "

Yes, this is exactly the "neighborhood" I'm thinking of (whether or not Mary's room is useful to me or not).

"Nagel takes a slightly different approach. In an effort to make his argument more adaptable and relatable, he takes the stand of humans attempting to understand the sonar capabilities of bats. Even with the entire physical database at one's fingertips, humans would not be able to fully perceive or understand a bat's sonar system, namely what it is like to perceive something with a bat's sonar."

This is another good example, although again not very useful in and of itself. This is the "type" of thing I'm thinking about though, one might comparing it to Human "feelings". Now, it's easy for a "math minded" person to casually discount or handwave away an individual person's "feelings" as some sort of a delusion, mental flaw, lack of education, etc. This is my natural tendency as well. But the older I get, the more I am beginning to think that this thinking will eventually be discovered to be incorrect. Something seems missing to me.

> but I think that the evidence that we have from biology and from artificial intelligence suggests that we could probably explain all of our abilities just with physical properties that we already understand

Oh sure, that's fine. I see that as the underlying mechanism, which is also important, but I'm not interested in that aspect of it. The resulting consequences, the nature and classification/understanding/deconstruction of consciousness, or more specifically quantifying human behavior, is all I'm interested in gaining a better understanding of.

I can't imagine the specific idea I have in mind is novel in any way, but whether anyone has considered it interesting enough to pursue (among all the other competing interesting ideas in this space) is what I'm trying to figure out.

This Mary's Room discussion may not be the exact topic I'm looking for, but it references the sorts of things I'm interested in.

So, it seems our choices are:

Qualia https://en.wikipedia.org/wiki/Qualia

The term qualia derives from the Latin neuter plural form (qualia) of the Latin adjective quālis (Latin pronunciation: [ˈkʷaːlɪs]) meaning "of what sort" or "of what kind" in a specific instance like "what it is like to taste a specific orange, this particular orange now". Examples of qualia include the perceived sensation of pain of a headache, the taste of wine, as well as the redness of an evening sky. As qualitative characters of sensation, qualia stand in contrast to "propositional attitudes",[1] where the focus is on beliefs about experience rather than what it is directly like to be experiencing.

Philosopher and cognitive scientist Daniel Dennett once suggested that qualia was "an unfamiliar term for something that could not be more familiar to each of us: the ways things seem to us"

Physicalism https://en.wikipedia.org/wiki/Physicalism

In philosophy, physicalism is the ontological thesis that "everything is physical", that there is "nothing over and above" the physical,[1] or that everything supervenes on the physical.[2] Physicalism is a form of ontological monism—a "one substance" view of the nature of reality as opposed to a "two-substance" (dualism) or "many-substance" (pluralism) view. Both the definition of "physical" and the meaning of physicalism have been debated.

Epiphenomenalism https://en.wikipedia.org/wiki/Epiphenomenalism

Physicalism x -1 -->

Pluralism https://en.wikipedia.org/wiki/Pluralism_(philosophy)

Pluralism is a term used in philosophy, meaning "doctrine of multiplicity", often used in opposition to Monism ("doctrine of unity") and Dualism ("doctrine of duality"). The term has different meanings in metaphysics, ontology, epistemology and logic. In metaphysics, Pluralism is the doctrine that - contrary to the assertions of Monism and Dualism - there are in fact many different substances in Nature that constitute Reality.

--> Substance Theory https://en.wikipedia.org/wiki/Substance_theory

Substance theory, or substance attribute theory, is an ontological theory about objecthood, positing that a substance is distinct from its properties. A thing-in-itself is a property-bearer that must be distinguished from the properties it bears.[1]

--> Thing-in-itself https://en.wikipedia.org/wiki/Thing-in-itself

Kant argued the sum of all objects, the empirical world, is a complex of appearances whose existence and connection occur only in our representations.[2] Kant introduces the thing-in-itself as follows: And we indeed, rightly considering objects of sense as mere appearances, confess thereby that they are based upon a thing in itself, though we know not this thing as it is in itself, but only know its appearances, viz., the way in which our senses are affected by this unknown something.

It's very easy for someone in no background in philosophy to get lost in a rabbit hole of interesting definitions, and then along the way you completely forget why you're doing this in the first place.

So, let's start over from a different angle...

Love https://en.wikipedia.org/wiki/Love

Love encompasses a variety of different emotional and mental states, typically strongly and positively experienced, ranging from the most sublime virtue or good habit, the deepest interpersonal affection and to the simplest pleasure.[1][2] An example of this range of meanings is that the love of a mother differs from the love of a spouse differs from the love of food. Most commonly, love refers to a feeling of strong attraction and emotional attachment.[3] Love can also be a virtue representing human kindness, compassion, and affection, as "the unselfish loyal and benevolent concern for the good of another".[4] It may also describe compassionate and affectionate actions towards other humans, one's self or animals.[5]

Alexithymia https://en.wikipedia.org/wiki/Alexithymia

Alexithymia /ˌeɪlɛksəˈθaɪmiə/ is a personality construct characterized by the subclinical inability to identify and describe emotions in the self.[1] The core characteristics of alexithymia are marked dysfunction in emotional awareness, social attachment, and interpersonal relating.[2] Furthermore, people with alexithymia have difficulty in distinguishing and appreciating the emotions of others, which is thought to lead to unempathic and ineffective emotional responding.[2] Alexithymia occurs in approximately 10% of the population and can occur with a number of psychiatric conditions.[3]

Frisson https://en.wikipedia.org/wiki/Frisson

Frisson (French for 'shiver') is a sensation somewhat like shivering, usually caused by stimuli other than cold. It is typically expressed as an overwhelming emotional response combined with piloerection (goosebumps). Stimuli that produce a response are specific to the individual.

http://www.slate.com/articles/health_and_science/science/201...

Listening to emotionally moving music is the most common trigger of frisson, but some feel it while looking at beautiful artwork, watching a particularly moving scene in a movie, or having physical contact with another person. Studies have shown that roughly two-thirds of the population feels frisson.

Where I'm going from here is basically, "So what?".

Roughly, the important idea I'm trying to get at is: what is the significance of all these ideas. Frisson is a particularly interesting phenomenon, in that different people feel it in response to very different things, and some people literally can't feel it at all!

Now take love....very similar phenomenon, but even more powerful, by far. Also, it is experienced vastly differently by (and, like Frisson, not felt by everyone).

So what?

Well, I think all of this is fundamental to the human experience, but more importantly, it is a fundamental root cause of why people can't get along (and recently, really really really can't get along).

So what?

Well, I wonder if maybe we can go a long way to fixing this problem of societal disunity if these ideas could somehow be summarized and effectively communicated to people. I think lots of people would be understanding/accepting of the premise, and lots wouldn't. Some would be hostile in various ways, and for various reasons (some honest and understandable, some evil). Of those who understand and appreciate the principle though, how many would "really get it". Sure, they could appreciate on an intellectual level, but "really getting it" is on the level of frisson and love. Is there a way to get people to that higher level? Why yes....yes there is.

I am going to push save now in case someone is still here and sees where I'm going with all this, and may know of any resources along this same theme.

To be continued....

"reality, at least as far as humans perceive it, consists of observations of physical events, or at least we know that much for sure."

Curious: would you say that your dreams are observations of physical events? Can you be sure that this isn't a dream? All we know is that we have experiences, and that it's a reasonable guess that something called "physical reality" is behind those experiences.

> would you say that your dreams are observations of physical events?

Current understanding is that dreams are a side effect of variational model selection. See Hobson, J. A., & Friston, K. J. (2012). Waking and dreaming consciousness: Neurobiological and functional considerations. Progress in Neurobiology, 98(1), 82–98. https://doi.org/10.1016/j.pneurobio.2012.05.003

> brain scans of subjects under these conditions find anything interesting

It is certainly possible, but currently not practical. Imaging technologies have limited spatiotemporal resolutions and coverage areas. Direct electrode reading are too intrusive, and so on. There is limited understanding of which brain areas might contribute to certain complex behavior so we might have to measure entire brain's activity. We don't know what resolution is useful - each spike or firing frequency or only dynamics (changes), etc. If we assume the availability of whole brain scan at spike resolution over extended time periods, theoretically a regression type algorithm could figure out relationship between brain activity and behavior. Firmly out of currently technology's reach though.

> definition/enumeration of self-awareness

Philosophers have been working on that for a long time. Some terms are well settled now, even though a term might cover a range of concepts. See that for "consciousness" here - https://plato.stanford.edu/archives/spr2018/entries/consciou.... This addresses mostly the western thought process and leaves out the eastern concepts. Most eastern concepts are too vague to pin down, though some are concrete enough to provide some insights (e.g. http://heavenmeetsearthyoga.com/news-features/the-meaning-of... matches reasonably well with modern understanding).

Is this based on common sense or intuition, or are you actually familiar enough with the current state of the art AI to say for certain that we actually do need the level of precision you refer to to get anywhere? Of course, the higher the precision the better, but do you know we can't get anywhere without it?

Current state of the art precision has yielded insights at the sensory-motor (e.g. precise understanding of what sensory inputs cause what neural activity) and at subsystem level (e.g. what mental activity "uses" what brain subsystems). I am sure more can be done with existing technology, but a grand unified theory would probably need more precision. It is somewhat akin to attempts at finding the grand unified theory of Physics that works both at quantum and cosmological scales.

I wonder if I did a poor job asking the question. What I'm getting at is, I wonder if common & reproducible brain patterns that might not formerly have been noticeable (perhaps due to a lack of precision when observing them in isolation) might reveal themselves when running assisted deep learning on a large number of subjects' results across a broad but "strictly defined" set of mental experiences that are known to produce "extreme psychological reactions". Might we discover associations (at the brain wave level) between experiences where our current understanding of the brain would not suggest there might be any? Would this type of analysis even need high precision?

Out of curiosity, was it you that downvoted my question? I'm struggling to understand what has changed on HN lately, where purely unopinionated and relevant questions are now very commonly downvoted. It's this type of thing that has caused my curiosity in this area, I'm interested in finding an underlying reason why people increasingly seem to be offended or disagree so strongly with things that only a few years ago were generally considered completely innocuous?

Perhaps it was the tone of the question? But again, I don't think that would have garnered any downvotes on HN 5 years ago - what's changed? Might we be able to detect something in brain activity that could lead us to some new paths to study to explain this widespread behavior?

Possibly related, but not necessarily:

https://en.wikipedia.org/wiki/Overfitting

https://www.quora.com/In-machine-learning-is-more-data-alway...

The book The Elephant In The Brain touches on this. Short version: effective self-deception is the only way to effectively deceive others, as we as a species have developed very finely tuned bullshit detectors. To recognize and acknowledge your own hidden motives is to ensure that you won't accomplish your evolutionary goals (survival/reproduction), so evolution has generally limited people's ability to gain insight into themselves.

It is entirely unremarkable that people should put "feeling" in italics in order to emphasize it, because it is precisely the thing gets handwavy non-explanations (like yours) from people who refuse to acknowledge the deep mystery (or "puzzle", if you want).

What is remarkable, to the point of comedy, is the reliability with which this refusal is accompanied by one or two sentence explanations that imply a miraculous end to the long, complex philosophical exploration of the mind-body problem that humanity has struggled with for millennia.

"They are processes in an organism, and we construct models of the self to help us get through the day" is one such meaningless explanation. You've answered nothing. You've done nothing. The problem of qualia? The unity problem? The knowledge problem?

The comparison with gravity was indeed unfair, but not in the way you think. It is unfair in that the problems of phenomenology and consciousness are much harder (see Knowledge Problem) because of their resistance to modeling (or any other kind of study that doesn't involve introspection).

To call "consciousness" and "feeling" folk terms and use that to dismiss incredibly simple and clear questions is the intellectual equivalent of tapping out. It's so simple: "what is the nature of the feelings themselves, apart from the dynamics of the system they are correlated with?"

Say we had a very detailed world simulation running on a massive server farm, like you see in some SciFi stories. Would the characters inhabiting this world have conscious experiences? They consist of complex bit patterns, or more accurately, electricity.

If you think they would, then let's take it one step further. A billion Chinese people are tasked with writing out the 1s and 0s on paper to implement the world simulation. This may take generations and the felling of forests, but it's done.

Would the process of writing down that program result conscious experiences the same as if the program was run on digital computers?

What about when the task is done? Would the paper be conscious or the ink on it? What if all the papers were separated by thousands of miles?

If you've read Greg Egan's Permutation City, then a similar sort of questioning was explored with the uploaded brain copies. The author ends up with his own weird "dust theory of consciousness", because there really isn't a good explanation for why any physical state would be conscious.

>Say we had a very detailed world simulation running on a massive server farm, like you see in some SciFi stories. Would the characters inhabiting this world have conscious experiences? They consist of complex bit patterns, or more accurately, electricity.

I think yes. I find the arguments for functionalism convincing.

After this you're rehashing Searle's Chinese room. As I've said, 'conscious experience' is not a properly defined concept, so I can only answer for what my own impressionistic judgement of its meaning is. For me, the entire system of people writing could maybe be described as having an experience if it directly encoded causally connected perceptions and actions of an agent in the virtual world. The whole thing is so far from how we use the word 'experience' that it is, as Dijkstra said, as interesting as asking whether a submarine can swim --- that is, not interesting at all.

Of all thought experiments, I've found the billion workers communicating with 1 and 0's to be the strongest illustration of the mind/body problem. It is just absurd at that point to believe consciousness could reside there.

I've wondered, too, whether there is a connection between the mind/body problem and the problem of existence (why is there something rather than nothing). These two problems seem to be fundamentally insurmountable.

> It is just absurd at that point to believe consciousness could reside there.

You’re just begging the question. ‘It sounds absurd’, is not a cogent argument.

I gave my intuition. There is no argument here.

"It is just absurd at that point to believe consciousness could reside there."

At the risk of losing my junior philosopher's secret decoder ring, I feel the need to point out that most philosophical writing is an elaborate logical ediface erected in support of the writer's previously existing, unreasoned prejudices.

Ok, can you explain if my laptop has consciousness?

Does it like to work for me, or hate it?

At what point will the first question be true?

And how can we construct machines that find working for us pleasurable?

Any reasonable definition of consciousness should IMO be a continuum, so your laptop has some consciousness (not very much).

I don't think concepts of like or hate are meaningful for the consciousness of your laptop. These kinds of responses probably arise from the kinds of goals that are created for agents by processes of natural selection.

We evolve them. We already have, they're called dogs.

> These kinds of responses probably arise from the kinds of goals that are created for agents by processes of natural selection.

You're speaking of "response", but it is supposed to be a concept internal to the laptop. Please be more clear, as the topic demands it.

Anyway, how do I create a laptop that finds pleasure in its work? Do I throw away the ones that don't work, and keep the ones that work? Is that enough selection?

And once I have a laptop that experiences some pleasure, why does it experience that amount of pleasure and not, say, ten times that amount?

I meant that the internal state of like/hate would be a response to something that the laptop experienced.

>Please be more clear, as the topic demands it.

You are using terms like 'pleasure' and 'experience' that have no scientific definition. I find your entire line of thinking unclear, but it doesn't help the tone of the argument to demand clarity.

>Anyway, how do I create a laptop that finds pleasure in its work? Do I throw away the ones that don't work, and keep the ones that work? Is that enough selection?

As with dogs, the internal state of the laptop would need to be malleable per generation, with generations of reproduction under adaptive pressures that favour the human owner. We currently don't have laptops of this kind.

> You are using terms like 'pleasure' and 'experience' that have no scientific definition.

We are not in the domain of science, but in the domain of philosophy ...

Also, don't you understand what I mean by "pleasure" or "experience"? Are you a bot? :)

> As with dogs, the internal state of the laptop would need to be malleable per generation, with generations of reproduction under adaptive pressures that favour the human owner. We currently don't have laptops of this kind.

We have genetic algorithms, so I'm sure we could come up with such laptops.

> We have genetic algorithms, so I'm sure we could come up with such laptops.

It is almost certain that we someday will. I doubt I'll still be alive to see it, though.

> Any reasonable definition of consciousness should IMO be a continuum, so your laptop has some consciousness (not very much).

That doesn't seem much different from the theory of consciousness Chalmers put forward in the second part of his 1996 book, where some law of nature connects consciousness to rich streams of information. That way, any physical system which creates a complex enough flow of information would be conscious to some degree.

The difference is that Chalmers recognizes that information and subjective experience are not equivalent, and thus you need a law of nature connecting the too. That was his proposal for expanding science.

I'm not saying whether he's write or wrong, only that Chalmers recognizes the hard problem can't be waved away by making an equivalence claim.

> The difference is that Chalmers recognizes that information and subjective experience are not equivalent,

"recognizes" seems to assume that he's right.

AFAICT there's absolutely no reason a priori to think that it's true, so I think the burden of proof is on those making the claim that the subjective is somehow not just the result of physical processes -- which is what the claim seems to amount to, at least to me.

Not only that, but as social creatures we have evolved mental models of other people, what we expect them to do, and say, and how we expect them to react to us, which in turn means we need to have models of our own behaviour in relation to theirs.

Do plants have a sense of self or references to its own "body" (my instinct tells me no, but honestly I have no idea). They interact with the world and stimuli around them much like animals do, albeit on a slower time frame.

Reply on news.ycombinator.com