On Consciousness

[PixyMisa]

Yet your posting your own personal view based on your own personal experience now

No. No, he's not.

 

[steenkh]

But they have to be. If you see blue when they see red, then at some level of abstraction the signals get switched.

Please note that both persons see the same colour. But nobody can tell if it looks the same inside their heads, but they both agree that they see "red". As I pointed out, since the brain patterns cannot be the same (the brains are different), they necessarily will have different experiences. Probably not a switching of colour experience, but more likely that they each experience something that the other does not experience at all. But both see the same colours.

 

[PixyMisa]

Please note that both persons see the same colour. But nobody can tell if it looks the same inside their heads

That's my point. We can tell. To get a different outcome, something must change at some point in the visual perception pathway, and since we know a lot about how the pathway works, we can probe the various stages to find out if they're working normally.

It's not a black box between the retina and subjective perception, but a series of progressively abstract representations processed in specific ways in specific locations in the brain.

 

[AlBell]

Wrong.

Turing test: fail. Code needs tweaking.

The question "does it look (feel) to me like it does to others?" is discussed in the Inverted Spectrum^WP argument, which has intrigued me for a long time.

My version has been: If I saw blue where others saw red, and vise versa, since infancy, how would that manifest itself to the outside world? How could that even be detected? How would even I know this?

Thanks for link. Thank FSM Hofstadter solves the problem so it doesn't exist.

Yet descriptions of an 'experience' are not the experience itself.

 

[PixyMisa]

Turing test: fail. Code needs tweaking.

Still wrong.

Thanks for link. Thank FSM Hofstadter solves the problem so it doesn't exist.

He doesn't solve the problem so much as point out that, like qualia and P-zombies, it's only a problem at all if your position is logically inconsistent to start with.

 

[Dancing David]

But it is the point. For subjective perception to end up dramatically different, at some point, something different happened in the processing of the sensory input.

And since each stage of the visual perception pathway has its own quirks, by examining those quirks we could identify where the change happened. Does the McCollough Effect tint vertical lines for our subject with the colour of a cherry or the colour of the sky?

The question, which as yet is unanswered is this:

Do I see cyan the same as you or do I see the hue you see as magenta or yellow?

We will agree on the color but the perceptual hue may be different.

 

[PixyMisa]

The question, which as yet is unanswered is this:

Do I see cyan the same as you or do I see the hue you see as magenta or yellow?

No, that question has been answered.

We will agree on the color but the perceptual hue may be different.

We can determine that by tracing the visual perception pathway, because that's where this happens. As I've noted, it's not a black box. The processing happens in specific ways in specific parts of the brain, and if it differs significantly from one person to another, it is possible to find that out.

 

[Belz...]

Yet your posting your own personal view based on your own personal experience now

Would you mind clarifying this word salad ?

 

[Belz...]

It's not a black box between the retina and subjective perception

And yet how many people would like it to be a black box...

 

[Belz...]

The question, which as yet is unanswered is this:

Do I see cyan the same as you or do I see the hue you see as magenta or yellow?

Why does it matter ?

 

[Dancing David]

Why does it matter ?

Which is what I asked AlBell

 

[Fudbucker]

I get the impression that some of you never read or simply ignore the posts that answer all those points of yours, and then act as though they were never posted.

If consciousness is computational, a simulation of consciousness is conscious because computation is what's required.

Again, we have unclear language. What do you mean by this? Consciousness is a KIND of computation? Consciousness is EQUIVALENT to computation? And what do you mean by "computational"?

COMPUTE
1: to make calculation

2: to use a computer

Neither of these are even close to satisfactory definition of what consciousness is.

I'm not persuaded much by computational theory. It's intriguing, but there's no science to back it up. The advocates here simply assert that it's true. The terms that we are familiar with by our own introspection (feel, think, be conscious) become trivialized to such an extent that they lose all meaning- if a mercury thermostat can feel (which has been asserted), then the word has nothing to do with what goes on when I stub my toe- the subjective experience of pain.

But there are some good points to be brought up:

1. Real photosynthesis and simulated photosynthesis are both clear terms. One is a physical process that produces energy, the other is a representation of that process and doesn't produce energy.
Real consciousness is not as clear a term, but we know it's a physical process of the brain (what else could it be? Anything else is dualism). Physical processes can be simulated, therefore consciousness can be simulated. But when you simulate a physical process, you have something that is different than the process being simulated. How is simulated consciousness different than real consciousness? It should be, if it's a physical process produced by a material thing, but it's not clear what simulated consciousness even is.

2. If a sufficiently complex computer can become conscious (no reason why it can't), then machine consciousness would obviously be computational in nature. That doesn't mean that organic consciousness is also computational, but it would evidence that it is. Unfortunately, we'll never know if machines achieve consciousness, so this evidence will always be elusive.

 

[trebor]

Real consciousness is not as clear a term, but we know it's a physical process of the brain..

Is a running computer program a 'physical process' of the computer?

I will open this question to the floor for comments as I do not seem to be getting any answers to any of my other questions

 

[rocketdodger]

So you are relying on a successfully passed Turing test as complete proof your computer is human level conscious. As I suspected. Thanks.

Yes.

However, it is worth noting that relying on a successfully passed Turing test would probably disqualify a number of humans themselves from being "human level" conscious. I don't think there is an issue with this, for instance I don't believe that an individual with severe mental handicaps is "human level conscious." That doesn't mean they don't see red, though.

Agreed that coding could differentiate the correct wavelengths we 'experience' as red. And none of us can ever verify how others 'experience' red either.

I'm not sure what you mean, here.

My meaning was just that if a machine is sitting here talking to you about how it experiences red, and it sounds like the same kind of stuff you would say, then you have no reason to believe that it is "faked" or "trickery" or any type of "chinese room" thing going on. The simplest explanation is that the machine does indeed experience red similarly to us. Otherwise, why would it be using similar language?

 

[rocketdodger]

Compare all the reporting you want to: None of it gets to the heart of the situation; 'what it actually feels like to me vs anyone else'.

Correct. But mathematically it is impossible to get to the heart of that situation, at least fully.

For example, suppose you want to know what it feels like to be a bat. Ignoring the fact that figuring out how to reconcile the two neural networks is monumental, try to imagine what the experience would actually entail.

Bat's don't have an understanding of human memory, so bat-you probably wouldn't be able to make sense of the memory that you were a human just seconds before. But bat-you doesn't care because it isn't thinking logically about anything besides bat desires, like finding food or whatever. You certainly wouldn't be capable of realizing "hey, I'm a bat now."

And when it is over -- what would your memories be? As human-you, how can you access bat-you feelings and perceptions? Your human brain has no body map for the bat body ( some parts might be similar, but others aren't ) and your human brain certainly can't hear the high frequencies that the bat brain can hear, so what would those memories be like? Just fuzzy grey areas in the past? If that is the case, then you still don't know what it feels like to be a bat, because you can't accurately remember much of it.

Knowing what it is like to be me, if you are you, is of course a lesser extreme, but the same problems remain. Fundamentally, the connectivity between neurons in my brain is the deterministic result of my past -- nothing else. Since we don't share the same pasts, there is no guarantee that any of the same connectivity can just be transplanted wholesale and retain meaning for the brain. In fact I highly doubt it would be the case for things like memory.

It seems more plausible to map your brain to some intermediate and see if that intermediate can be re-mapped to my brain, then maybe I could experience your past memories *as if* it was me experiencing those events in the first place. But then that isn't really knowing what it is like to be someone else !! That is just sort of "being in their shoes."

 

[rocketdodger]

Please note that both persons see the same colour. But nobody can tell if it looks the same inside their heads, but they both agree that they see "red". As I pointed out, since the brain patterns cannot be the same (the brains are different), they necessarily will have different experiences. Probably not a switching of colour experience, but more likely that they each experience something that the other does not experience at all. But both see the same colours.

I think it would only be accurate to say that people see "different" colors, or even "different" hues, if everything post-visual-processing in their brain was the same.

But as you point out, that is never the case. So saying different people see things differently doesn't even really mean anything, it is literally as cryptic as saying a tree sees things differently from an asteroid.

It makes more sense to speak in terms of what goes on in the brains of two people who are seeing red is similar, or dissimilar, or sort of similar, or whatever, and the ways things are similar, or dissimilar, or sort of similar, etc, are this: < then include the stuff Pixy has been talking about>

 

[Fudbucker]

Is a running computer program a 'physical process' of the computer?

I will open this question to the floor for comments as I do not seem to be getting any answers to any of my other questions

What else could it be?

 

[Belz...]

Again, we have unclear language. What do you mean by this? Consciousness is a KIND of computation? Consciousness is EQUIVALENT to computation? And what do you mean by "computational"?

I'm not playing this time. This focus on semantics is useless.

Cars are vehicles.

 

[Belz...]

Correct. But mathematically it is impossible to get to the heart of that situation, at least fully.

For example, suppose you want to know what it feels like to be a bat. Ignoring the fact that figuring out how to reconcile the two neural networks is monumental, try to imagine what the experience would actually entail.

Bat's don't have an understanding of human memory, so bat-you probably wouldn't be able to make sense of the memory that you were a human just seconds before. But bat-you doesn't care because it isn't thinking logically about anything besides bat desires, like finding food or whatever. You certainly wouldn't be capable of realizing "hey, I'm a bat now."

And when it is over -- what would your memories be? As human-you, how can you access bat-you feelings and perceptions? Your human brain has no body map for the bat body ( some parts might be similar, but others aren't ) and your human brain certainly can't hear the high frequencies that the bat brain can hear, so what would those memories be like? Just fuzzy grey areas in the past? If that is the case, then you still don't know what it feels like to be a bat, because you can't accurately remember much of it.

Knowing what it is like to be me, if you are you, is of course a lesser extreme, but the same problems remain. Fundamentally, the connectivity between neurons in my brain is the deterministic result of my past -- nothing else. Since we don't share the same pasts, there is no guarantee that any of the same connectivity can just be transplanted wholesale and retain meaning for the brain. In fact I highly doubt it would be the case for things like memory.

It seems more plausible to map your brain to some intermediate and see if that intermediate can be re-mapped to my brain, then maybe I could experience your past memories *as if* it was me experiencing those events in the first place. But then that isn't really knowing what it is like to be someone else !! That is just sort of "being in their shoes."

It's even worse than that. It's even difficult to remember what it felt to be human-you just a few years ago.

 

[Beelzebuddy]

As loathe as I generally am to engage someone who resorts to dictionary attacks so early, you seem a fairly reasonable person so I'll give it a shot.

Again, we have unclear language. What do you mean by this? Consciousness is a KIND of computation? Consciousness is EQUIVALENT to computation? And what do you mean by "computational"?

"Consciousness" refers to an instance of the particular set of computations performed by our brains, the way your OS (while running) is an instance of a particular set of computations. Or for a more direct analogy, the way the rocket guidance of the Saturn V was an instance of the particular set of computations built into its analog computer. Unfortunately we don't have a good word for an instance of a particular set of computations that isn't easily confused with the set's products or the algorithms the computations are generated from.

The terms that we are familiar with by our own introspection (feel, think, be conscious) become trivialized to such an extent that they lose all meaning- if a mercury thermostat can feel (which has been asserted), then the word has nothing to do with what goes on when I stub my toe- the subjective experience of pain.

It has everything to do with it. How do you think pain receptors sense injury? The physical processes are very similar, and this is the most vital part of the pain response. While the way thermostats feel heat is a pale shadow of our own, at the most basic level it's the same thing - a system sensing its environment and reacting appropriately.

I'll admit this is largely hyperbole, but it's there for a reason: making an anthropocentric distinction right off the bat tends to cloud everything that follows in a kind of low grade dualism. You can track the pain response all the way up and down, hitting nothing but physical processes performing minor computation the whole time, and there'd still be people arguing "Yes, but that's not feeeling."

1. Real photosynthesis and simulated photosynthesis are both clear terms. One is a physical process that produces energy, the other is a representation of that process and doesn't produce energy.

It must produce simulated energy, or it's not accurately-simulated photosynthesis. But I think you're presuming consciousness to be the energy in this analogy. It's not. It's the photosynthesis itself, the instances of that particular set of computations. Whether those computations are performed by physical macromolecules or simulated ones is inconsequential, the process is the same. They're both photosynthesis.

2. If a sufficiently complex computer can become conscious (no reason why it can't), then machine consciousness would obviously be computational in nature. That doesn't mean that organic consciousness is also computational, but it would evidence that it is. Unfortunately, we'll never know if machines achieve consciousness, so this evidence will always be elusive.

We've yet to find anything in the brain that can't be described as a computational process. Some people posit that we eventually will, a hypothesis that some here (myself among them) derisively call the "magic bean" theory. It's a god of the gaps for dualism: consciousness is the product of physical computation plus a magic bean, without which the computation alone could never be conscious.