On Consciousness

[!Kaggen]

Of course I am biased, being a programmer, but IMO it is the good programmers who understand more about our world than anyone else.

Needing to write your own code that numerically simulates/models a world tends to make you think about just exactly what that whole "world" thing entails.

I think it is the exact opposite.
You know more and more about less and less when you reduce the world to numbers.

 

[AlBell]

No.

None of these people you speak of really know that much about the subject. Baars knows a little more, but not by much, since he isn't a computer scientist.

The people that know alot about the subject are the researchers making robot brains out of neural networks. Brains that function using the same principles as ours.

You wrote quite a bit of stuff, but none of it is really relevant, because there are researchers making robot brains out of neural networks that function using the same principles as ours.

I repeated that, because it is important.

I missed the demonstration that you or anyone else actually knows what principles our brains use.

Blathering SRIP just doesn't fill the gap at least for me.

 

[tensordyne]

See Bolded: Sort of agree with this, from my purely scientific side, in terms of physically measurable experimental evidence being the essential precursor of any theory.

See italisized: And note purely the emphasis.

I honestly dont know how I can make you see my side of the argument without advocating doing something I don't like recommending to people unless they make that decision themselves.

Sorry Mr. Scott, I would go into why I think consciousness is a fundamental aspect of the universe (albeit I am pushing my guessing ability some here) but I have seemingly pissed off someone who is raring to go into everything about neural networks and hard AI and so on, so let me just say that Zeuzzz's response above aint too bad.

Zeuzzz, are you referring to giving DMT a try? It is a fascinating subject this molecule DMT. I have not tried it myself. I want to though (even though it sounds scary as hell).

 

[tensordyne]

I missed the demonstration that you or anyone else actually knows what principles our brains use.

Uhmmm, yeah, that is about what I was going to say.

 

[tensordyne]

No.

None of these people you speak of ...

Wow, I find the above quote in its full form very odd in a way. Let me see if I can break it down some.

No.

Not exactly the best way to start a conversation.

None of these people you speak of really know that much about the subject. Baars knows a little more, but not by much, since he isn't a computer scientist.

What subject is it that we are speaking off? The subject of neural nets, consciousness (the main point of this thread), neurobiology, ...??? So far I have used the following researchers names': Dennett, Chalmers, V. S. Ramanujan, Searle. I don't think I brought up Baars ever, not that I super care. I am just trying to be complete.

I have to say, even though I disagree with Dennett, he does know a lot about the literature and concepts of the topics I listed above as to subject matter. It is pretty silly to assume he doesn't; the guy is super smart.

Similar things can no doubt be said about Chalmers and perhaps less so for Searle. As for Ramanujan, the hard on you seem to have for programmers knowing so much is not true. Most programmers do not know much about Brain Physiology, even if they are experts in programming neural nets. I could go on about domain specific knowledge. Instead, I will leave that for a response to a different entry.

The people that know alot about the subject are the researchers making robot brains out of neural networks. Brains that function using the same principles as ours.

Ah, I see that the subject is neural nets (?). The current paradigm is that the brain mostly acts like a neural net. That model is still subject to change as we learn more.

Myself, I see a definite possibility that neurons might be acting like little antennae as well as being rapid firing circuits. Some indication of this to me is due to the fact that consciousness has been found to be associated with neurons firing synchronously.

If the brain is only a net, I do not see the point behind neurons firing synchronously (consciousness seems to especially coincide with synchronous firing in disparate parts of the brain, or so I remember reading somewhere).

The structure of neurons also makes me think they may be acting as antennae. They often have long axons coated in segments as well as fire off in rapid bursts. Many man-made antenna have similar characteristics. Look up the papers by Johnjoe McFadden (and perhaps some others) if you want to read from someone who knows much more than I do about this particular subject.

I have to say, when thinking about a brain that is wired up like a switching circuit (neural network, I am being metaphorical calling it a switching circuit) versus one that has both switching and acts like a field effect computer (not that we even have such a technology, but it does make sense as a concept perhaps), the latter one allows for much more complex interactions. I say this as both someone who knows a fair amount of physics and computer science.

Modeling a Faraday cage encased set of rapidly switching antennae would be pretty costly computationally. Especially if the order of antennae was even anywhere near as many neurons as a small brain has. Lots of nonlinear EM field effects would be present. It is just a mess of a problem.

In any eventuality, it is premature to exclude out all aspects of the brain except neural networks as a point of study in relation to consciousness. What about various chemicals? Do they not affect consciousness? I am sure others can add in various aspects of the brain that do not function like neural nets and could very easily be connected to conscious states.

Plus, neural nets are not all they are cracked up to be. They have limitations and their own issues (please feel free to gift us with your knowledge on this rocketdodger).

You know, if you asked someone during the Victorian times how the brain works they might relate it to a steam engine. In our day we relate it to electronic computers. Who knows what next technology we will relate it to that is supposedly the truth of the matter.

So I say, let the researchers in their various fields investigate. I have no problem with that (even AI research!). I do have a problem if someone comes along and claims their invention is conscious. Let the neuroscientists do their job to hopefully figure out how consciousness works in the one source we can seemingly all agree about: us.

You wrote quite a bit of stuff, but none of it is really relevant, because there are researchers making robot brains out of neural networks that function using the same principles as ours.

Well, that is somewhat galling to say that what I wrote is irrelevant on the grounds above. The only thing that seems irrelevant is the response you gave to the post I made. Irrelevant, that is, to anything I wrote (note that I returned the disfavor in kind, I am just not that rude), because there is not much in the way of analysis of what I said. I do not have any problem with anyone bringing neural nets into the mix though, so that is cool.

The response also seemed a bit canned. Almost like how a religious person will repeat their testimony ad nauseum to anyone who would care to listen. "Neural Networks, Brains, Programmers, Go...". "Jesus, Heaven, Church, Go..."

I repeated that, because it is important.

No thanks, you do not need repetition for me to acknowledge a concept. I say this for future reference. Of course, if you could acknowledge a few of the concepts I bring up by deigning to respond to them, that would be most appreciated.

 

[rocketdodger]

I think it is the exact opposite.
You know more and more about less and less when you reduce the world to numbers.

That doesn't really make sense.

To understand X implies the ability to reduce X into simpler concepts and see how those simpler concepts work together to produce the behavior of X.

Repeat that, and eventually you arrive at "reducing the world to numbers."

If you mean that looking at the brain in a narrow sense like this implies no knowledge of things the brain can do such as art and poetry, well I would agree. But I don't see how knowing about art and poetry means one knows "more" about the brain.

 

[rocketdodger]

I missed the demonstration that you or anyone else actually knows what principles our brains use.

Blathering SRIP just doesn't fill the gap at least for me.

I know you did, and that is funny, because it was in a post directed at you, a post you actually responded to BTW.

Par for the course, though. Leave out the maths and hard stuff, lets get back to arguing about stupid red herrings, eh?

 

[tensordyne]

Blathering SRIP just doesn't fill the gap at least for me.

Sorry, this just bothers me, what the heck does SRIP stand for? "Soldier Robot Interface Project" is the best I could find online, so I at least did some searching before asking.

 

[rocketdodger]

Not exactly the best way to start a conversation.

I am not starting. You are starting. I have been in the current conversation, with the same people, for about 4-5 years.

I have to say, even though I disagree with Dennett, he does know a lot about the literature and concepts of the topics I listed above as to subject matter. It is pretty silly to assume he doesn't; the guy is super smart.

The people I consider authorities on the subject are the programmers that are actually doing the research trying to model the brain. Everyone else just has an opinion, and you know what they say about opinions.

Ah, I see that the subject is neural nets (?). The current paradigm is that the brain mostly acts like a neural net. That model is still subject to change as we learn more.

This is a very silly group of statements, given that the brain is made of neurons that are networked. It doesn't "act like" a neural network, it IS a neural network,

Myself, I see a definite possibility that neurons might be acting like little antennae as well as being rapid firing circuits. Some indication of this to me is due to the fact that consciousness has been found to be associated with neurons firing synchronously.

If the brain is only a net, I do not see the point behind neurons firing synchronously (consciousness seems to especially coincide with synchronous firing in disparate parts of the brain, or so I remember reading somewhere).

The structure of neurons also makes me think they may be acting as antennae. They often have long axons coated in segments as well as fire off in rapid bursts. Many man-made antenna have similar characteristics. Look up the papers by Johnjoe McFadden (and perhaps some others) if you want to read from someone who knows much more than I do about this particular subject.

I have to say, when thinking about a brain that is wired up like a switching circuit (neural network, I am being metaphorical calling it a switching circuit) versus one that has both switching and acts like a field effect computer (not that we even have such a technology, but it does make sense as a concept perhaps), the latter one allows for much more complex interactions. I say this as both someone who knows a fair amount of physics and computer science.

Modeling a Faraday cage encased set of rapidly switching antennae would be pretty costly computationally. Especially if the order of antennae was even anywhere near as many neurons as a small brain has. Lots of nonlinear EM field effects would be present. It is just a mess of a problem.

In any eventuality, it is premature to exclude out all aspects of the brain except neural networks as a point of study in relation to consciousness. What about various chemicals? Do they not affect consciousness? I am sure others can add in various aspects of the brain that do not function like neural nets and could very easily be connected to conscious states.

Well, let me just ask you a few things.

1) Do you have any actual evidence of these things, or just speculation based on how neurons look similar to "other stuff," such as octopi and trees, in pictures?

2) Can you even begin to suggest why these things you speak of might be necessary for consciousness, that is, what the neural network alone could not manage?

3) Can you even begin to describe how these things you mention might even work, in the context of the neural circuitry of the brain that we know quite a bit about? Merely saying "use field effects" isn't exactly a description.

And BTW chemicals affect how the neural network functions, so I don't see how various drugs and chemicals could advance any argument that it isn't all about neural networks.

Plus, neural nets are not all they are cracked up to be. They have limitations and their own issues (please feel free to gift us with your knowledge on this rocketdodger).

Last time I checked, so do human brains.

Well, that is somewhat galling to say that what I wrote is irrelevant on the grounds above. The only thing that seems irrelevant is the response you gave to the post I made.

But everything you said is irrelevant, given the fact that researchers are making robot brains out of neural networks, that function using the same principles as our brain.

You are basically telling the wright brothers that they need magic faerie dust to fly, after they already got the plane in the air. Yeah they don't have a fighter jet, but the plane is flying.

20 years ago your theories might have been more relevant, but they just aren't anymore.

The response also seemed a bit canned. Almost like how a religious person will repeat their testimony ad nauseum to anyone who would care to listen. "Neural Networks, Brains, Programmers, Go...". "Jesus, Heaven, Church, Go..."

More like how a scientist will repeat known research ad nauseum to people who claim such research doesn't exist.

Sorry it exists. It exists. It exists. I don't know how else to dress up that information. It exists.

No thanks, you do not need repetition for me to acknowledge a concept. I say this for future reference.

Ok, but I don't really care if you merely acknowledge a statement of mine. I hope you would actually *think* about it, and try to absorb the information, rather than just rushing on to write a response.

Of course, if you could acknowledge a few of the concepts I bring up by deigning to respond to them, that would be most appreciated.

I did, when you answer the questions I posed we can get deeper into it.

 

[tensordyne]

Of course I am biased, being a programmer, but IMO it is the good programmers who understand more about our world than anyone else.

Needing to write your own code that numerically simulates/models a world tends to make you think about just exactly what that whole "world" thing entails.

Programmers have a certain set of skills. I guess you could say these skills are in the area of applied finite mathematics. Those skills do not give the programmer any preference to "understand more about our world than anyone else." I really doubt for instance you are going to come up with a new law of physics using string manipulation.

Programmers generally have to know much less about our world than do Medical Doctors. Most programmers that I have met really do not know much in the way of physics or chemistry beyond what you would expect an average person to know. In fact, I do not see any good reason why anyone should expect a programmer to know or get the "world" idea better than anyone else in any of the many disciplines out there. They are all just different ways of looking at things or doing things.

Perhaps though you are referring to the modeling mindset that programmers have. Well, many other disciplines share this same mindset too.

 

[Mr. Scott]

See Bolded: Sort of agree with this, from my purely scientific side, in terms of physically measurable experimental evidence being the essential precursor of any theory.

See italisized: And note purely the emphasis.

I honestly dont know how I can make you see my side of the argument without advocating doing something I don't like recommending to people unless they make that decision themselves.

It's not just purely scientific experimental evidence. There's no evidence of any type that consciousness exists anywhere except in brains on Earth.

I do see your side of the argument. I used to believe in universal consciousness, which I think is what you are advocating. Sorry, no evidence for it.

You don't have to advocate that I do something to get me to see your side of the argument. Just tell us what it is. If it's dangerous I won't do it, but it seems obvious that it will amount to an argument from personal experience, which would not constitute evidence of any sort. Testimony doesn't cut it.

 

[AlBell]

Sorry, this just bothers me, what the heck does SRIP stand for? "Soldier Robot Interface Project" is the best I could find online, so I at least did some searching before asking.

Self Referential Information Processing.

That, and read "I Am A Strange Loop" by Hofstadter, author of Godel Escher Bach.

None of it gets into the principles a brain uses to form consciousness.

ps. We have a poster who now writes conscious programs.

 

[Mr. Scott]

ps. We have a poster who now writes conscious programs.

linky please?

 

[quarky]

An ant has a neural network, and an ant colony behaves like a meta neural network...even using chemicals to transmit information between units.
Is it safe to say that an ant colony is conscious?
How about the individual ant?

Is a network of humans conscious, beyond the consciousness of the individuals within?
Can an organization take on a life of its own, with its own agenda?
Is the Supreme Court conscious, for instance?

 

[tensordyne]

Part I.

I am not starting. You are starting. I have been in the current conversation, with the same people, for about 4-5 years.

Fantastic, it still does not change the fact that a simple "No" at the beginning of a post is ... I don't know, not conveying a lot except general disapproval. It seems kind of childish to me too, but perhaps that is just me.

The people I consider authorities on the subject are the programmers that are actually doing the research trying to model the brain. Everyone else just has an opinion, and you know what they say about opinions.

Please indicate subject matter we are speaking of. I had issues with this before. I was looking for "... on the subject of xxx...".

This is a very silly group of statements, given that the brain is made of neurons that are networked. It doesn't "act like" a neural network, it IS a neural network,

The neural networks as you use them in a computer programming context are very different than the tissues of the brain. The brain is not just composed of neurons, and yet neural networks (computer-wise, AKA, artificial neural networks) do not address the space between the neurons. Or the exacting physics or chemistry of the brain. Or who knows what else. Most likely lots and lots of stuff (cell dynamics???).

Neural Networks are a model of one aspect of the brain. That is why they were invented as a concept in the first place. Neural Networks (computer science version) also usually do not have anywhere near the complexity and feedback that the networks of neurons, glial cells, etc. have in a real brain.

So no, your brain does not act like an artificial neural network and also is not an artificial neural network. Your brain has a network of neurons, it is not a neural network (if anything, the network connections are at times graph-like, at other times mostly like an artificial neural network, and at other points like long distance telephone wires.).

Simplifying the brain only down to neural networks does not seem right and proper in my book unless the dynamics you are trying to capture fits well within that model. I remain unconvinced that the Neural Network concept either encapsulates the main functioning of the brain or in any way helps to
figure out how consciousness operates.

Well, let me just ask you a few things.

By all means.

1) Do you have any actual evidence of these things, or just speculation based on how neurons look similar to "other stuff," such as octopi and trees, in pictures?

Do not underestimate the investigative power of something looking like something else. Of course to really test this antenna idea you would have see if it could recieve or send signals. As far as I know, this has not been done yet (I would do it myself. Chances of me getting published even if I did
show it are nill is my guess. Plus, large octopi neurons are not easy to come by! I am not a biologist either...).

We do know that the brain responds to EM fields and that it even produces a Field of its own. The brain having its own field is how EEG works.

Electroencephalography (EEG) is the recording of electrical activity along the scalp. EEG measures voltage fluctuations resulting from ionic current flows within the neurons of the brain.

When you have current flow you get - magnetic fields. I trust I do not have to explain how an electric field would come about in this type of situation (hint, ionic molecules of various kinds). That is just physics.

So, given the above, it seems pretty plausible to me that neurons are antennae. It is my pet theory, so take it or leave it. Even then, I do respect science, so I am not even going to completely believe it myself until it is printed up in a Journal somewhere (with positive follow up papers hopefully!).

2) Can you even begin to suggest why these things you speak of might be necessary for consciousness, that is, what the neural network alone could not manage?

Well, I hope you do not mind me going down the rabbit hole a bit on this one. You have asked me to guess, so I feel that there should be a certain amount of latitude in my response. I will understand if you do not find the ideas compelling. This question involves the basis of 'consciousness', so I have to readily admit anything I say now about that is at best educated speculation (that is true to the best of my knowledge for everyone on the planet).

That said, I think I know what consciousness is about. I do not know its actual basis. I guess I can say the following at least. If consciousness is of physical EM field origin, then while it may be useful as part of your greater model of the brain to include neural network elements, a neural network is not an EM field. Fields are defined for each point in a given space. The two mathematical structures are very different.

Here is the rabbit hole part. Perhaps our brains are like radio stations that pick up channel "consciousness". Inputs from sensing organs modulate the consciousness signal so that one experiences various sensations based upon various stimuli.

Yeah, I know. It is a bit out there. On the other hand, even if the above paragraph is wrong, the physical basis of consciousness could just as easily still be EM (other?). I go with the evidence whatever the case may eventually turn out to be. So far the evidence is that the brain is electrical and has EM fields. These are two essential facts of the situation.

3) Can you even begin to describe how these things you mention might even work, in the context of the neural circuitry of the brain that we know quite a bit about? Merely saying "use field effects" isn't exactly a description.

Very true what is said above. The hallmark of a good analytic mind is to ask "how". To give some idea of how I think an animal brain would work in this schema, perhaps it is best to give a description of how I would perhaps invent an EM field effect brain myself.

First I would make a Faraday Cage the size of the brain (if you look up McFadden you can see the paper on CEMI where he explains how our brains are encased in a kind of Faraday cage). Perhaps to make this interesting I would hook up some wires to some form of sensor. To go with the robot theme maybe I would have output wires controlling a motor or something.

Next I would make a bunch of small antennae. Each antennae would be controlled by a central circuit (think neuron soma) of its own. There are many types of neurons so perhaps I would not just use antennae that only have one pole. In general though most of the antennae would have one large antennae part (think Axon). The antennae I would have would also have the ability to send out one or more direct wire signals to other antennae as a further means of inter-antennae communication (here think of Axon Terminals connecting to Dendrites).

Perhaps I would use a genetic algorithm to figure out how to hook up the antennae together. Or perhaps I would just try and approximate the general features of a given animal brain (a simple one at that). Each neuron/antennae would also be given the task of trying to figure out when it will receive a fire signal (defined appropriately) and try and fire just before that (maybe include some psuedo-random firing just to be safe).

Who knows, this would be research, so to a certain extent, one would just have to play around and see what happens (play in this sense would still have to be following the Scientific Method!). Each one of the parts and steps in creating the 'brain' above comes from actual physiological facts about biological brains (just like how neural network ideas came from the physiology of the brain). I hope this gives a decent picture of what I am thinking about.

 

[tensordyne]

Part II.

And BTW chemicals affect how the neural network functions, so I don't see how various drugs and chemicals could advance any argument that it isn't all about neural networks.

OK, lets set things straight then. This is getting crazy because it feels like each paragraph based response is worthy of an essay all its own. You can take a square peg and if it is small enough fit it into a round hole. Something very similar is going on here.

When talking about neural networks we can either be referring to the artificial or biological kind. If we are talking about biological then drugs will most likely be a consideration of any experimental tissue model. Even then, tissue models are just that, models. They do not include other cells the brain has nor quite likely are they anywhere near as complex as real brain tissue is. If it was as complex there would not be much point in using the experimental tissue model.

If we are talking artificial neural networks and how they advance the argument that it is not all about neural networks, then we have an even worse case. A brain's neurons generally do not line up as perfectly in layers like an artificial network does. There is no well defined input, hidden or output layer. The topology is much more graph-like. Plus, network models (or graph models), only deal with connections.

Say you wanted to model the Electromagnetic Vector Field with neural nets. It does not make a lot of sense to do so. Or maybe you want to model various scalar fields such as chemical densities. Again, this does not work well or at all with neural nets. Or perhaps let's try and do some organic chemistry with neural networks. No, that does not make sense either. Yet, each of those aspects of the brain is no doubt important.

In general, you need more than one tool to analyse a real brain. All of those tools are not neural nets either. Square meet circle.

Last time I checked, so do human brains.

Interesting point. I was actually teasing out to see what you knew about the limitations of neural networks as a model of computation. See the quote by Dewdney below to get a taste of what I am referring to.

But everything you said is irrelevant, given the fact that researchers are making robot brains out of neural networks, that function using the same principles as our brain.

What do you mean by 'brains'. Brains with the same capabilities as animal brains? Sorry, I do not find that to be at all likely. Most of the circuits we create, whether neural network or not, confine electrons to certain paths and damp out EM fields. The physics of any robot "brain" I know of works in quite a few ways differently than an animal brain does. This is a fact.

You are basically telling the wright brothers that they need magic faerie dust to fly, after they already got the plane in the air. Yeah they don't have a fighter jet, but the plane is flying.

Nope, I am saying we have not figured out consciousness yet. That the people who will figure it out are the neuroscientists (with help from whatever fields are relevant). I am saying that my best guess, given all that I have read, is that consciousness has a physical basis (no magic faerie dust needed).

I hate to have to ding you because I know that you are using figures of speech to make your point. It is just that being able to restate accurately someone else's position is a large part of being intellectually honest. For instance, I think the Intelligent Design people are off their rocker. If I wanted to talk about them though I am going to use their own words, not a jestful recapitulation of what I think their position is like.

Please, at any time if I misquote you are otherwise misstate your concepts, let me know immediately. I do not want to be arguing against paper tigers.

20 years ago your theories might have been more relevant, but they just aren't anymore.

Because 20 is when computational neuroscience really started taking off? Consider the following criticism I found through a simple Wikipedia search.

A. K. Dewdney, a former Scientific American columnist, wrote in 1997, "Although neural nets do solve a few toy problems, their powers of computation are so limited that I am surprised anyone takes them
seriously as a general problem-solving tool." (Dewdney, p. 82)

Odd that. Seems like neural nets might not be all they are cracked up to be.

More like how a scientist will repeat known research ad nauseum to people who claim such research doesn't exist.

There exists tons of research on all sorts of topics. I hope that was never at issue. Understanding the brain as primarily being a neural net is a good idea seems to me to be one of your main messages. If such is the case, I am not convinced of this.

I have read too much to know that the dynamics of neural nets is does not adequately cover how the brain functions. Additionally, I do not see how neural nets leads to consciousness either in principle or in some theoretical sense.

Sorry it exists. It exists. It exists. I don't know how else to dress up that information. It exists.

So do neuroscience journals that almost never talk much about neural nets ever. I think you have really grafted onto this idea of neural nets as being so important that it is a good idea to review why. I have seen this kind of behavior before. When Fractals were all the rage then everything was about fractals. Fractals, fractals everywhere.

Your models are too limiting. Being a physics minded person, if I had to model the brain as completely as I could maybe I would use a lattice model. Some other models or combination of known models would be my approach. There are lots of ways to model the brain. Neural nets are just one of many type of models. Most likely they are not even one of the most important ones.

Ok, but I don't really care if you merely acknowledge a statement of mine. I hope you would actually *think* about it, and try to absorb the information, rather than just rushing on to write a response.

Oh, I have thought pretty hard on these subjects. Neural Nets are not new to me. I am pretty sure that maybe some of the things I have said are new to you though (I for instance would be interested in why you think we do not have to understand the Biology of Consciousness before we can recreate it artificially, if I understand your position correctly. Everything else that we recreated from Biology meets that requirement, I just do not see why we can skip past that requirement.). Either way, it is all good. We are not likely to solve it in a few pages in this format.

As far as acknowledgement, no, but to skip wholesale as you did and then jump right into neural nets without even a lead up, well what do you want? It seemed kind of rude to me. Like you are basically saying, "naw, those other subjects do not matter".

Why? Not because you showed them to be really irrelevant in their given context. It was because in your own mind the only thing that is important is talking about robots controlled by neural nets or something.

I am sorry, I am not very impressed by neural network controlled robots. I find the subject to be worthy of study. It has helped in understanding the bigger picture of the universe. I wonder though, would you claim that because these robots have neural nets they are conscious? If so, you are going to have to do a lot more work to convince me of that supposed fact.

I did, when you answer the questions I posed we can get deeper into it.

Very well. I appreciate your response.

Just a last quick word. I do not think rocketdodger that you understand what is meant by the word 'consciousness' as used when referring to what it is like to be something (what some call phenomenal consciousness). I think abstraction has occluded the reality of the concept of experience (or degraded it to the point that everything must be stated in terms of mental models).

I see this in Dennett too. I am not saying this to be disparaging. It kind of makes me sad in a way when someone, possibly because of prior training or just turn of mind, can not seem to grasp an important concept.

The world turns on though.

 

[tensordyne]

Self Referential Information Processing.

That, and read "I Am A Strange Loop" by Hofstadter, author of Godel Escher Bach.

None of it gets into the principles a brain uses to form consciousness.

ps. We have a poster who now writes conscious programs.

Very good. Thanks for letting me know.

 

[Beelzebuddy]

The neural networks as you use them in a computer programming context are very different than the tissues of the brain. The brain is not just composed of neurons, and yet neural networks (computer-wise, AKA, artificial neural networks) do not address the space between the neurons. Or the exacting physics or chemistry of the brain. Or who knows what else. Most likely lots and lots of stuff (cell dynamics???).

Neural Networks are a model of one aspect of the brain. That is why they were invented as a concept in the first place. Neural Networks (computer science version) also usually do not have anywhere near the complexity and feedback that the networks of neurons, glial cells, etc. have in a real brain.

So no, your brain does not act like an artificial neural network and also is not an artificial neural network. Your brain has a network of neurons, it is not a neural network (if anything, the network connections are at times graph-like, at other times mostly like an artificial neural network, and at other points like long distance telephone wires.).

Simplifying the brain only down to neural networks does not seem right and proper in my book unless the dynamics you are trying to capture fits well within that model. I remain unconvinced that the Neural Network concept either encapsulates the main functioning of the brain or in any way helps to
figure out how consciousness operates.

The search term you should be using is neuronal networks.

Do not underestimate the investigative power of something looking like something else. Of course to really test this antenna idea you would have see if it could recieve or send signals. As far as I know, this has not been done yet (I would do it myself. Chances of me getting published even if I did
show it are nill is my guess. Plus, large octopi neurons are not easy to come by! I am not a biologist either...).

We do know that the brain responds to EM fields and that it even produces a Field of its own. The brain having its own field is how EEG works.

When you have current flow you get - magnetic fields. I trust I do not have to explain how an electric field would come about in this type of situation (hint, ionic molecules of various kinds). That is just physics.

So, given the above, it seems pretty plausible to me that neurons are antennae. It is my pet theory, so take it or leave it. Even then, I do respect science, so I am not even going to completely believe it myself until it is printed up in a Journal somewhere (with positive follow up papers hopefully!).

Well, I hope you do not mind me going down the rabbit hole a bit on this one. You have asked me to guess, so I feel that there should be a certain amount of latitude in my response. I will understand if you do not find the ideas compelling. This question involves the basis of 'consciousness', so I have to readily admit anything I say now about that is at best educated speculation (that is true to the best of my knowledge for everyone on the planet).

That said, I think I know what consciousness is about. I do not know its actual basis. I guess I can say the following at least. If consciousness is of physical EM field origin, then while it may be useful as part of your greater model of the brain to include neural network elements, a neural network is not an EM field. Fields are defined for each point in a given space. The two mathematical structures are very different.

Here is the rabbit hole part. Perhaps our brains are like radio stations that pick up channel "consciousness". Inputs from sensing organs modulate the consciousness signal so that one experiences various sensations based upon various stimuli.

Yeah, I know. It is a bit out there. On the other hand, even if the above paragraph is wrong, the physical basis of consciousness could just as easily still be EM (other?). I go with the evidence whatever the case may eventually turn out to be. So far the evidence is that the brain is electrical and has EM fields. These are two essential facts of the situation.

Very true what is said above. The hallmark of a good analytic mind is to ask "how". To give some idea of how I think an animal brain would work in this schema, perhaps it is best to give a description of how I would perhaps invent an EM field effect brain myself.

First I would make a Faraday Cage the size of the brain (if you look up McFadden you can see the paper on CEMI where he explains how our brains are encased in a kind of Faraday cage). Perhaps to make this interesting I would hook up some wires to some form of sensor. To go with the robot theme maybe I would have output wires controlling a motor or something.

Next I would make a bunch of small antennae. Each antennae would be controlled by a central circuit (think neuron soma) of its own. There are many types of neurons so perhaps I would not just use antennae that only have one pole. In general though most of the antennae would have one large antennae part (think Axon). The antennae I would have would also have the ability to send out one or more direct wire signals to other antennae as a further means of inter-antennae communication (here think of Axon Terminals connecting to Dendrites).

Perhaps I would use a genetic algorithm to figure out how to hook up the antennae together. Or perhaps I would just try and approximate the general features of a given animal brain (a simple one at that). Each neuron/antennae would also be given the task of trying to figure out when it will receive a fire signal (defined appropriately) and try and fire just before that (maybe include some psuedo-random firing just to be safe).

Who knows, this would be research, so to a certain extent, one would just have to play around and see what happens (play in this sense would still have to be following the Scientific Method!). Each one of the parts and steps in creating the 'brain' above comes from actual physiological facts about biological brains (just like how neural network ideas came from the physiology of the brain). I hope this gives a decent picture of what I am thinking about.

No. No to all of this. No to each and every sentence. No.

 

[tensordyne]

The search term you should be using is neuronal networks.

Good point. I do not see that it makes much difference in terms of the analysis already given though.

No. No to all of this. No to each and every sentence. No.

A lazy if complete response. I am sorry that I have insulted your religious sensibilities. I only say that because, in my experience, someone does not at least give a reason for disagreement only when one touches upon certain tenets of that person's faith. Since these tenets of faith can not really be logically defended, the best response is just a repeated 'no'. Just a guess.

On the other hand, perhaps I should just respond to the argument and not the arguer. I would if I could. No arguments were given. This is a classic fallacy BTW. It is called "refusal to reason".

Let me know if you want to engage in actual debate that involves reasoned arguments at some point in the future.

 

[blobru]

--- First I would make a Faraday Cage the size of the brain (if you look up McFadden you can see the paper on CEMI where he explains how our brains are encased in a kind of Faraday cage). ...

Johnjoe McFadden's CEMI summary page (scroll down for links to three of his articles).