Has consciousness been fully explained?

[rocketdodger]

Still, all of these different flavors of consciousness are produced by brains, and we have every reason to believe that the mechanism will be analogous for all conscious animals.

You mean like running, or flying?

Hmmm... I wonder if those things can be done via non-biological mechanisms ...

Speculating about some other form of consciousness, or a "general consciousness" that includes such forms, is empty wordplay.

Well I suppose that's how epistemology works, piggy. Sorry, I didn't make the rules.

You look at a bunch of stuff, you determine what that stuff has in common, and you label it. Then, if there is no reason other stuff couldn't have similar attributes, you have no choice but to admit it. You can't just stick your head in the sand. I mean, you can, but it is pretty irrational.

Thats like saying hey, horses and people and ostriches run because of muscle cells, so anything without muscle cells can't run.

No -- those things run because of the way their supporting limbs move them along. So quite obviously anything else that has supporting limbs that move it along in a similar fashion might be said to "run." Hence a general theory of running.

 

[rocketdodger]

I don't even know what is meant by "program the neuron's behaviour" in this context. And we can't program _all_ the neuron's behaviour, obviously. And the brain is not a closed system. And so on.

Now, if we produce a series of copies of the brain, each one more and more like the brain, then probably, when we get close enough, we'll have something that is effectively like a brain. But we really don't know what that point is.

Rational people are pretty sure that if you make a copy of the brain at per-particle resolution you will have something that is effectively "like" a brain.

So yeah, we kinda do.

 

[rocketdodger]

The trouble comes when folks mistake the abstractions for physical reality and come to absurd conclusions, such as believing that a "tinker toy brain" could perform all the functions of a human brain.

Oh, so your position is that all the functions of the human brain are necessary for consciousness?

But you just said that many animals are also conscious ...

Your arguments display much inconsistency.

 

[rocketdodger]

Well, it would certainly be nice if it turned out that you and I were talking about the same thing in different terms and with different frames of reference.

I kind of doubt this, tho, b/c of some of the conclusions which your point of view leads you to, if I remember correctly.

But perhaps I do not remember correctly.

You didn't understand either my point of view nor the conclusions, back when I stated them, so even if you did remember correctly your memory is of stuff that is incorrect.

I wouldn't worry about judging arguments by the conclusion, I would worry about the actual premises and logic used to reach the conclusion.

You can't just accept an argument and then bail when the conclusion is something you find distasteful. If you accept the argument you have to accept the conclusion. Thats what we call "rational thought."

 

[rocketdodger]

Humans are born conscious, as far as we can tell.

And in any case, that's not an answer to the question of why certain neural activity in the brain is involved in generating conscious experience, but most of it is not.

If you genuinely think that a newborn infant is conscious, then you better start re-thinking your approach.

Because I can tell you right now that the similarities between a newborn's brain and the brain of even a 2 year old are extremely significant.

Furthermore the set of behaviors that a newborn is capable of is pretty small.

At this point, you are pretty close to agreeing with Pixy about the consciousness of a programmable thermostat.

Just saying ... you should think a little more about the implications of your views.

 

[PixyMisa]

Yeah, by my definition human infants are almost certainly conscious, though it's not easy to map what's going on in there. Just wonder what definition Piggy is using to make that statement.

 

[Malerin]

If you genuinely think that a newborn infant is conscious, then you better start re-thinking your approach.
Because I can tell you right now that the similarities between a newborn's brain and the brain of even a 2 year old are extremely significant.

Do you think newborns aren't conscious? That would certainly help your position

Furthermore the set of behaviors that a newborn is capable of is pretty small.

At this point, you are pretty close to agreeing with Pixy about the consciousness of a programmable thermostat.
Just saying ... you should think a little more about the implications of your views.

Suggest you read up on brain physiology of newborns.

 

[laca]

Trivially, the position and momentum of every particle in the neuron, for example.

It is not necessary to know the position and momentum of every particle in a system in order to be able to program its external behavior. By the same reasoning, nothing can be programmed, not even a ball dropping to the ground. Which is of course utter nonsense. Your argument is invalid. Try again?

 

[Belz...]

I don't even know what is meant by "program the neuron's behaviour" in this context.

I think I meant the same thing you did when you said it.

And we can't program _all_ the neuron's behaviour, obviously.

Why not ?

And the brain is not a closed system.

Neither is the simulation.

 

[PixyMisa]

It is not necessary to know the position and momentum of every particle in a system in order to be able to program its external behavior.

Right. That can't be required because (a) the precision to which one of those properties is even defined depends on the other and (b) if the brain were that finely balanced there would be no life on the planet more complex than an amoeba.

It's not an argument, it's just an excuse.

 

[rocketdodger]

Do you think newborns aren't conscious? That would certainly help your position

I think that, on a sliding scale, the consciousness of newborns is much more similar to that of programmable thermostats than to adult humans.

Suggest you read up on brain physiology of newborns.

Go ahead and gimme a citation if you can find one. Cmon, do it. Or just a link to some wikipedia article backing up your point. Anything.

I suggest *you* read up on neural development. In particular

http://en.wikipedia.org/wiki/Synaptic_pruning

http://en.wikipedia.org/wiki/Synaptic_plasticity

That you think the topography of a newborn's neural networks are relatively similar to that of even a 2 year old is indicative of your knowledge on this issue.

 

[laca]

The fundamental flaw in the computationalist reasoning is the assumption that the very limited digital representation of neural activity is the only thing that's important in what's going on in the brain. That's not a justified assumption.

There is lots of evidence that it is and none that it isn't. Why would one say it's not a justified assumption?

 

[PixyMisa]

And more to the point, what else might be required? I've yet to see a coherent response to that question.

 

[westprog]

It is not necessary to know the position and momentum of every particle in a system in order to be able to program its external behavior. By the same reasoning, nothing can be programmed, not even a ball dropping to the ground. Which is of course utter nonsense. Your argument is invalid. Try again?

At some point in this discussion, people started using the phrase "program its behaviour". I don't really know what this means. Does it mean "perform all the functions of the system" or "provide a simulation of the way the system works" or what?

It's possible to do a simulation of a system, which will provide insight into how the system behaves - but it's never a perfect, or complete simulation, and it will never predict precisely how the system will behave. So it's possible to do a very simple simulation of a ball dropping to the ground, using just initial position and gravity, but for a more accurate simulation, you'd need to program in the shape of the ball and air resistance. That would be more accurate, but you'd also need to allow for wind, temperature, etc etc.

Of course, the most important point is that no matter how accurate your simulation becomes, there will never be any actual behaviour reproduced.

In the case of the behaviour of the brain, since we don't know what precise activity of the neurons produces consciousness, to claim that it can be simulated using some subset of the neuron activity is a prediction too far. And to claim that the simulation will actually produce the effect is simply begging the question. In order to believe that consciousness can be produced by computer operations, one has to believe that consciousness is effectively produced by computer operations - something we certainly don't know.

 

[westprog]

There is lots of evidence that it is and none that it isn't. Why would one say it's not a justified assumption?

"lots of evidence"? I don't agree that there's "lots of evidence". I think that there's evidence that some of what the brain does appears to work in a way analagous to a digital network - but it's an analogy, not a circuit diagram. As of now, it's not possible to replace even one neuron with a digital electronic substitute. Do that and you might have something.

 

[westprog]

Neither is the simulation.

In what sense is the simulation not a closed system? Because a Turing machine is a closed system, by definition.

 

[laca]

At some point in this discussion, people started using the phrase "program its behaviour". I don't really know what this means. Does it mean "perform all the functions of the system" or "provide a simulation of the way the system works" or what?

Inputs are transformed into outputs. We know how neurons do that. Given the same inputs, we can produce the same outputs without involving a neuron in the process.

You still haven't managed to show why step 1. in my argument is flawed.

 

[laca]

"lots of evidence"? I don't agree that there's "lots of evidence". I think that there's evidence that some of what the brain does appears to work in a way analagous to a digital network - but it's an analogy, not a circuit diagram.

It doesn't have to be. That's why it's analogous and not the same.

As of now, it's not possible to replace even one neuron with a digital electronic substitute. Do that and you might have something.

Who said anything about replacing neurons in a brain with digital parts? The whole argument revolves around the idea that if we can emulate the behavior of a neuron, in theory we can emulate a whole brain, and as a corollary, consciousness as well.

Of course, since the neuron is so different from a digital circuit, my bet is that if/when humanity manages to create a consciousness, it won't be by following that path.

 

[westprog]

It doesn't have to be. That's why it's analogous and not the same.



Who said anything about replacing neurons in a brain with digital parts? The whole argument revolves around the idea that if we can emulate the behavior of a neuron, in theory we can emulate a whole brain, and as a corollary, consciousness as well.

It's the if that is at the heart of it. Then if we get past the if, we have to consider whether the emulated consciousness is the same as consciousness.

Of course, since the neuron is so different from a digital circuit, my bet is that if/when humanity manages to create a consciousness, it won't be by following that path.

I agree. The assumption that a neuron is a simple digital switch is not justified on by close examination.

 

[Piggy]

Right. Thanks.

No need for the rolleyes.

Suppose you asked what part of a power plant can't be programmed.

A: The power plant.

When you're talking about behavior generated by a coordinated set of physical objects operating in real spacetime, there is no pure-programming solution that will generate the same behavior.

And for that matter, what part of an atom can't be programmed?

A: The atom.