On Consciousness

[tsig]

And I pointed out I have given those tests and you don't know what you are talking about, I didn't see an answer then, nor now.

You have to ask a question to get an answer. All I've seen is assertions and insults,

 

[Jeff Corey]

But at what level of abstraction? The brain is a computer (a pulse-coded switched digital network) and the neurons are the logic gates, so when you get down to that level of detail, dealing with computation is unavoidable.

When you're looking at a broader picture, you can use black-box models for the various brain functions, and it works perfectly well. But when you drill down from the black-box models, or build up from the molecular level, in the middle it's a computer. (An odd, squishy, unreliable computer, but still...)


There are worse fates in life.

OK, I know that and have some idea how that can lead to us feeling subtlely different levels of input rather than y/n.
Not my area of expertise.
I'm into studying behavior that can be reliably measured. At this time we have no way of measuring what you're talking about with CAT scans, PET_CT scans (had one today, serious, all protein diet, no starch or sugar before) so I don't care about computer consciousness.

 

[PixyMisa]

Yes, I know; however, I followed up the Wiki quote: "The terms are not quite synonymous: "super-Turing computation" usually implies that the proposed model is supposed to be physically realizable, while "hypercomputation" does not", and found a lot of talk and papers about physical implementations of what are claimed to be super or supra-Turing neural nets (e.g. Neural & Super-Turing Computing), but also the paper I previously linked suggesting that at least one of these can be analytically reduced to a Turing-complete model.

I've had a chance to scan through those papers now. Both (actually, it's the same research both times) make the mistake of assuming that reality is infinitely divisible, or to be more precise, meaningfully infinitely divisible. And it's just not so. Even if you don't come down on the side of the Planck scale being an actual limit - of spacetime being strictly quantised - it is still an effective limit, and any realisable analogue computer is still only Turing-complete; the function of any finite analogue computer can be replicated by a digital computer.

 

[PixyMisa]

OK, I know that and have some idea how that can lead to us feeling subtlely different levels of input rather than y/n.
Not my area of expertise.
I'm into studying behavior that can be reliably measured. At this time we have no way of measuring what you're talking about with CAT scans, PET_CT scans (had one today, serious, all protein diet, no starch or sugar before) so I don't care about computer consciousness.

Yes. The computational model is all very well, but it's about as broad a theory as you can get and still be considered a theory at all. To be of any use it needs exactly the sort of hard practical work you're doing.

My dispute is with those who categorically say no, consciousness can't possibly be computation but can provide neither evidence nor reasoned argument to support their position. And with whoever took those Oriental Beef Stir Fry instant meals off the market, because those things were delicious.

 

[tsig]

Whether or not one can discriminate between two colors is not the same as what those colors actually look like. Maybe it doesn't matter. Different people have different favorite colors, but that might be more because of association than how their color qualia are manifesting for them.

Then what does determine what the colors actually look like?

 

[!Kaggen]

Yes. The computational model is all very well, but it's about as broad a theory as you can get and still be considered a theory at all. To be of any use it needs exactly the sort of hard practical work you're doing.

My dispute is with those who categorically say no, consciousness can't possibly be computation but can provide neither evidence nor reasoned argument to support their position. And with whoever took those Oriental Beef Stir Fry instant meals off the market, because those things were delicious.

Okay so Pixy Misa passed the Turing test for me.
What do you others think?

 

[dlorde]

...The perceptual 'map' of the visual field is abstracted is some ways and is certainly confabulated in the blind spot when looking through only one eye...

The confabulation is the point I was trying to make. I guess expressed it poorly - I was just contradicting the common misperception that there is a photograph-like neural representation of the visual scene in the visual cortex onto which the retinal information is mapped, and that missing details (e.g. blind spot) are filled in with scenic data as the eye saccades over the the scene. That doesn't happen.

As you suggest, in the single eye situation, this 'gap' is actually filled during subsequent processing by extrapolation/interpolation of the surrounding data - a 'more like this' approach - demonstrated by the spoked-wheel-with-hidden-hub image, where when the blind spot is positioned over the hidden hub, the spokes appear to extend to meet in the middle.

Sorry for the confusion.

 

[dlorde]

I'm thinking (and am aware of these thought and words, plus the taste of tomato soup in my mouth) that there needs to be a separate thread about the properties (or otherwise?) of "absolutely nothing".

What would we talk about?

 

[dlorde]

I've had a chance to scan through those papers now. Both (actually, it's the same research both times) make the mistake of assuming that reality is infinitely divisible, or to be more precise, meaningfully infinitely divisible. And it's just not so. Even if you don't come down on the side of the Planck scale being an actual limit - of spacetime being strictly quantised - it is still an effective limit, and any realisable analogue computer is still only Turing-complete; the function of any finite analogue computer can be replicated by a digital computer.

OK. That first paper was just a random sample, but seems representative. So I guess the physical neural net implementations said to be 'super-Turing' can be treated as convoluted instances of Turing complete systems, and the unimplemented ideas are either Turing complete or unrealisable 'hyper-computers'.

 

[Belz...]

Unfortunately, I have to make a living.

Translation "I have a life..."

 

[PixyMisa]

OK. That first paper was just a random sample, but seems representative. So I guess the physical neural net implementations said to be 'super-Turing' can be treated as convoluted instances of Turing complete systems, and the unimplemented ideas are either Turing complete or unrealisable 'hyper-computers'.

That's what it looks like to me. And for once, this is actually my field.

 

[Dancing David]

Are you saying consciousness is a brain process???

No conscious computers?

Did you read my comments in the thread? I do believe that computational consciousness is probable, however I do not myself define consciousness as computation. I believe that a process very similar to consciousness can come from computation and that the two would be equivalent.

However I personally believe that the brain functions would be better modeled by a giant analog array. The brain functions in ways that definitely can be modeled by computers. I myself do not define brain processes as computation.

I have stated many times that an adequate model would need an array of one trillion by two thousand, the messy nature of the feedback loops of the other receptors and dendritic connections would require further large arrays as would the modeling of the biological effects of the system.

I would say that brain processes are comparable to computation and could be modeled as such, and I see no reason why conscious computers can not be created. They would certainly be smaller than an analog array of the scale needed.

 

[Dancing David]

I'm unclear what qualifies as a "behavior called consciousness", but would you agree that passing the mirror test is a behavior called consciousness?

If one defines consciousness that way, it is an alleged behavior of consciousness. What I find funny is that there are people who belief dogs and cats are not conscious.

 

[Dancing David]

...could you possibly be any more obsequious?


Have not had time to respond to your verbal orgasm yet DD...but the time will come. Unfortunately, I have to make a living.

Thats fine, I do understand. While not currently working ( I am a school minion): I am very busy doing work on my house (knocking caulk from windows currently), caring for a relative and visiting another in the hospital.

I was over the top, however I await your explanation of consciousness as something more than biochemistry.

I do apologize for my level of snark, rudeness and confrontation.

 

[Dancing David]

Okay so Pixy Misa passed the Turing test for me.
What do you others think?

I think that these conversations have ranged for years on this forum. In particular in the R&P section there have been a wide variety of arguments and debated.

-brains as receivers
-qualia and consciousness as unobservable
-kantian metalogic
-the impossibility of objective study
-the special nature of qualia and consciousness
-the vague problem of definition (also called the hard problem of consciousness)
-special pleading
-privileged pleading
-defintional pleading
-pleading for some extra special nature of consciousness
-the universal nature of consciousness

The list goes on and on and on, usually coming down to this : consciousness is something that some people want to believe is different from everything else in the world. there is absolutely no evidence that it is, there is zero evidence that it is not a result of biological brain processes.


So after you participate in twenty or so of these slog matches, it is easy to get frustrated. There are times I can control my frustration but here is SMT I am dismayed by the philosophical pandering when it occurs.

 

[Dancing David]

The confabulation is the point I was trying to make. I guess expressed it poorly - I was just contradicting the common misperception that there is a photograph-like neural representation of the visual scene in the visual cortex onto which the retinal information is mapped, and that missing details (e.g. blind spot) are filled in with scenic data as the eye saccades over the the scene. That doesn't happen.

As you suggest, in the single eye situation, this 'gap' is actually filled during subsequent processing by extrapolation/interpolation of the surrounding data - a 'more like this' approach - demonstrated by the spoked-wheel-with-hidden-hub image, where when the blind spot is positioned over the hidden hub, the spokes appear to extend to meet in the middle.

Sorry for the confusion.

None asked for on my part I misunderstood what you were saying, I apologize if I was rude.

 

[dlorde]

None asked for on my part I misunderstood what you were saying, I apologize if I was rude.

No problem.

There is a noticeable increase in politeness and apology here of late - I wonder if it's connected to the possibility of an imminent improvement in the weather?

 

[Dancing David]

I could wish, the high today is 86, back to the 90s for another two weeks.

I was getting belligerent, which I do regardless of the weather.

 

[Zeuzzz]

After all these pages in this thread, what have we learned about consciousness?

That even the 'experts' here don't have a friggin clue. Try as they may.

Every specialist in each different discipline has approached this from very different perspectives.

Taking a step back and re-evaluating their tunnel vision might be in order for some of them.

This does belong in philosophy.

 

[Belz...]

That even the 'experts' here don't have a friggin clue. Try as they may.

That people disagree does not discount what they way.