My Model of Human Consciousness

[rocketdodger]

I'm still having a hard time understanding how one equates computation with conscious experience. You've provided a very good description of information processing and reasoning but none of your OPs address or explain how data translate into experience. You just assume that if a system contains and manipulates data that it has subjective experience of said data. The model you've presented completely ignores the salient features of our own BNN and assume that abstracted functions of it are essentially the same.

No matter how one slices it Symbol ≠ Meaning, Syntax ≠ Semantics, and Computation ≠ Consciousness.

If you can think of an aspect of your own experience that cannot be explained by a reasoning system such as I have described, I would be happy to discuss it. In fact I would really like to know, because it would mean my model is insufficient, and I care about that.

But this isn't the R&P section of the forum, so there won't be any metaphysical nonsense here. If you don't have specific ideas to discuss then don't bother posting anything at all.

 

[rocketdodger]

A fact is a true statement about the world.

That isn't a very useful description.

First, because the truth of a statement is relative to the owner of the knowledge base.

Second, because the term "statement" isn't useful unless it is further defined.

Third, because even once "statement" is defined, no entity carries around "statements" in it's knowledge base. Do you think statements about the world exist in your brain, prior to you speaking or writing?

The definition I gave is operational and is very useful. Because it implicitly contains the information that all facts are owner relative (they only exist in the knowledge base of an entity, and only the entity has access to its own knowledge base), and that this information exists in the form of a relation between two or more other pieces of information.

But if you really like the term "statement," I think you will find that most accepted definitions of "statement" are equivalent to some type of relation.

 

[rocketdodger]

This can be easily disproved. Clear your mind of everything but a pink elephant. You're not doing any reasoning, but you're conscious.

Again, thank you for not reading 90% of the OP before you responded -- something I specifically asked people not to do in first post of the thread.

So, in case you didn't bother to read anything else, let me repeat it for you -- the function of a neural network can be thought of as reasoning, where each impulse(s) of a neuron is a fact and the summation of incoming action potentials at the neuron bodies is a simple type of inference, the inference being that if enough impulses arrive within a certain time a new impulse is generated and sent out (if enough facts are true, a new fact is inferred).

 

[rocketdodger]

ftfy

My model easily explains qualia. In fact, qualia make more sense under my model than the models that champion qualia to begin with.

Qualia are just another type of reasoning, so your statement "I believe qualia is the root of human experience" is simply a stronger form of my own statement "I believe reasoning is the root of human experience."

 

[INRM]

I personally think it's completely unethical to create a simulation of a sentient being, even if it is educational to do so. I have explained why I felt this to be unethical in the past, though you may or may not agree with me.

INRM

 

[CapelDodger]

All very well thought out, rocketdodger, but it gets us nowhere with the really hard problem, which is where self-consciousness, self-awareness, the "me" that lives a few inches behind the bridge of my nose and looks out through my eyes, fits into science as we know it.

I don't buy into the supernatural for a moment, and solipsism is a pet-hate of mine, but I think there's a transcendent quality to the knowing experience of existence that can never really be explained. It can be modelled or mimiced, it might even be an emergent phaenomenon of some systems, but that can never bridge the gap between objective and subjective existence.

Which is why I've given up on it .

 

[AkuManiMani]

I have to disagree with that last part. It seems to me that the exact opposite is true; conscious experience is the root of human reasoning. Reasoning is something we do with our experiences -- reasoning itself doesn't actually explain experience, tho.

Well, you either didn't understand 90% of the OP or you posted this without reading 90% of the OP, because I make it clear exactly what I mean by "reasoning" and why it applies to machines as well as individual neurons, never mind entire human brains.

I understood your OP. Right off the bat you described facts as the relations between percepts [a term thats interchangeable with 'qualia'] and reasoning as the process of inferring previously unknown facts from those one already has.

I stick by my statement. What you're calling percepts [i.e. qualia] are the basis of conscious experience. Reasoning is simply what we do with conscious experience.

What you haven't established in you OP is that systems other than that of BNNs can actually have percepts. You just take for granted that if a system is manipulating 'symbols' then it must have some subjective experience of them.

If you can think of an aspect of your own experience that cannot be explained by a reasoning system such as I have described, I would be happy to discuss it. In fact I would really like to know, because it would mean my model is insufficient, and I care about that.

But this isn't the R&P section of the forum, so there won't be any metaphysical nonsense here. If you don't have specific ideas to discuss then don't bother posting anything at all.

Who even needs to get into metaphysics? I'd be satisfied with just physics at this point. So far, your model is just an abstracted description of neural net computation which assumes that any systems modeling such functions are conscious. My lymph nodes, and endocrine systems do not produce conscious experiences even tho they meet your functional criteria. Why should I assume that an artificial computer following such a model would produce it when we don't even know how it is produced by our own biology?

The best we can do right now is to speculate off of what little we do know and -- I'm genuinely sorry to say -- your assumption of percept = symbol doesn't really cut it

 

[rocketdodger]

I personally think it's completely unethical to create a simulation of a sentient being, even if it is educational to do so. I have explained why I felt this to be unethical in the past, though you may or may not agree with me.

INRM

lol, ok.

What do you think about the rest of the OP, besides that one sentence?

 

[rocketdodger]

All very well thought out, rocketdodger, but it gets us nowhere with the really hard problem, which is where self-consciousness, self-awareness, the "me" that lives a few inches behind the bridge of my nose and looks out through my eyes, fits into science as we know it.

It doesn't?

I think it does, but of course I can only speak for myself. My own self-awareness is merely a huge aggregation of symbols that are somehow related to myself. Is yours different?

It isn't that hard to think of various ways that a symbol for "self" might arise, given how from conception we are learning the differences between the part of the world that is self and the part that is non-self.

And if such a symbol existed, could a neural network not include it in a reasoning chain, just like any other symbol?

 

[rocketdodger]

I understood your OP. Right off the bat you described facts as the relations between percepts [a term thats interchangeable with 'qualia'] and reasoning as the process of inferring previously unknown facts from those one already has.

I stick by my statement. What you're calling percepts [i.e. qualia] are the basis of conscious experience. Reasoning is simply what we do with conscious experience.

What you haven't established in you OP is that systems other than that of BNNs can actually have percepts. You just take for granted that if a system is manipulating 'symbols' then it must have some subjective experience of them.

Erm, not really....

A percept is simply a direct piece of information about the world. They have nothing to do with qualia. The firing of a single neuron in your retina is a percept. So is the waveform of a microphone or the voltage of a simple photoelectric cell.

What you call "qualia" are not percepts, by defintion, because they are not direct. They exist at a higher level than sensory neurons.

Who even needs to get into metaphysics? I'd be satisfied with just physics at this point. So far, your model is just an abstracted description of neural net computation which assumes that any systems modeling such functions are conscious. My lymph nodes, and endocrine systems do not produce conscious experiences even tho they meet your functional criteria. Why should I assume that an artificial computer following such a model would produce it when we don't even know how it is produced by our own biology?

Again, if you want to discuss specific ideas, I am happy to do so. But there are no specifics in this last paragraph, so I am not sure how to respond.

 

[AkuManiMani]

It doesn't?

I think it does, but of course I can only speak for myself. My own self-awareness is merely a huge aggregation of symbols that are somehow related to myself. Is yours different?

It isn't that hard to think of various ways that a symbol for "self" might arise, given how from conception we are learning the differences between the part of the world that is self and the part that is non-self.

And if such a symbol existed, could a neural network not include it in a reasoning chain, just like any other symbol?

Remember that symbols are just associative triggers to evoke meanings in a subjects mind. I think your model provides a good architectural framework for the reasoning process but right now we don't know what physically constitutes percepts. Science might find the answer in our life time and it might not. You just gotta learn to deal with it in the meantime, man.

 

[rocketdodger]

Here is a good starting point if anyone wants references about the pattern detectors in the visual cortex I spoke of: http://en.wikipedia.org/wiki/Feature_detection_(nervous_system)

Turns out they are called "feature detectors," I had forgotten the name over the years!

 

[Malerin]

That isn't a very useful description.

First, because the truth of a statement is relative to the owner of the knowledge base.

Wha? A proposition is true if it corresponds to reality: "All ravens are black" is true if all ravens are black. It has nothing to do with how it relates to "the owner of the knowledge base". If "the owner of the knowledge base" believes all raves are white, no relationship between the facts of the owner's knowledge base will make that belief true. It will always be false.

Second, because the term "statement" isn't useful unless it is further defined.

A proposition is true if it accurately describes reality: "God exists" is true if God exists. "God does not exist" is true if there is no God.

 

[mimada]

Sorry to nitpick, but when you say:

The best way to understand this is to look at the difference between the memory required to store a font compared to the memory required to store an image of the alphabet.

I believe you mean vector font vs. bitmapped font.

 

[rocketdodger]

Sorry to nitpick, but when you say:



I believe you mean vector font vs. bitmapped font.

Yes I do, thank you.

 

[AkuManiMani]

Erm, not really....

A percept is simply a direct piece of information about the world. They have nothing to do with qualia. The firing of a single neuron in your retina is a percept. So is the waveform of a microphone or the voltage of a simple photoelectric cell.

What you call "qualia" are not percepts, by defintion, because they are not direct. They exist at a higher level than sensory neurons.

I thought by 'percept' you meant definition 1:

per⋅cept  [pur-sept]
–noun
1. the mental result or product of perceiving, as distinguished from the act of perceiving; an impression or sensation of something perceived.
2. something that is perceived; the object of perception.

If you're defining percepts are sensory data below the level of conscious awareness then you're model isn't really even a description of consciousness.

Who even needs to get into metaphysics? I'd be satisfied with just physics at this point. So far, your model is just an abstracted description of neural net computation which assumes that any systems modeling such functions are conscious. My lymph nodes, and endocrine systems do not produce conscious experiences even tho they meet your functional criteria. Why should I assume that an artificial computer following such a model would produce it when we don't even know how it is produced by our own biology?

Again, if you want to discuss specific ideas, I am happy to do so. But there are no specifics in this last paragraph, so I am not sure how to respond.

Thats just it. The specifics of consciousness aren't in yet. There is no scientific knowledge of what consciousness is, so any proposed theoretical models of it, at this point, are purely speculative. The only safe assumption right now is that conscious experience is something tied only to the activity of BNN.

You're trying to put the technical cart before the scientific horse, while completely foregoing the philosophical driver. Its not going to go anywhere.

 

[rocketdodger]

Wha? A proposition is true if it corresponds to reality: "All ravens are black" is true if all ravens are black. It has nothing to do with how it relates to "the owner of the knowledge base". If "the owner of the knowledge base" believes all raves are white, no relationship between the facts of the owner's knowledge base will make that belief true. It will always be false.




A proposition is true if it accurately describes reality: "God exists" is true if God exists. "God does not exist" is true if there is no God.

You are clearly not talking about anything that relates to this thread, so I am not even going to bother responding to the above.

If you are unable to get past the collision between the definition of "fact" that I am using and the definition you are locked in at, then you can call it "dofjadf" or "dkjjgjgj" or "mmmmsnnsre," without changing the content of the OP.

 

[rocketdodger]

I thought by 'percept' you meant definition 1:

per⋅cept  [pur-sept]
–noun
1. the mental result or product of perceiving, as distinguished from the act of perceiving; an impression or sensation of something perceived.
2. something that is perceived; the object of perception.

If you're defining percepts are sensory data below the level of conscious awareness then you're model isn't really even a description of consciousness.




Thats just it. The specifics of consciousness aren't in yet. There is no scientific knowledge of what consciousness is, so any proposed theoretical models of it, at this point, are purely speculative. The only safe assumption right now is that conscious experience is something tied only to the activity of BNN.

You're trying to put the technical cart before the scientific horse, while completely foregoing the philosophical driver. Its not going to go anywhere.

... if you want to discuss specific ideas I would be happy to.

 

[Malerin]

You are clearly not talking about anything that relates to this thread, so I am not even going to bother responding to the above.

If you are unable to get past the collision between the definition of "fact" that I am using and the definition you are locked in at, then you can call it "dofjadf" or "dkjjgjgj" or "mmmmsnnsre," without changing the content of the OP.

[fingers in ears] nyah nyah nyah I can't hear you [/fingers in ears]

 

[rocketdodger]

[fingers in ears] nyah nyah nyah I can't hear you [/fingers in ears]

Why are you putting your fingers in your ears?