As far as I can see there are two possibilities of how consciousness relates to the brain:

1. That it is the mirror image of processing in the physical neural networks of the brain and is therefore superfluous. It could just as easily be there as not because all the 'calculations' are performed by the underlying physical hardware.
2. It is more of a 'black box' that is an actual processing element in the brain and therefore 'interfaces' with the physical hardware in some way. In its other plane of existence, whatever that may be, it can perform processing that cannot be achieved by the neural networks alone. In this idea if (some of) the neural networks of the brain were essentially relegated to the roles of input and memory only, their data would be 'set' by processing in this other plane and a hypothetical examination of the brain would show changes in value seemingly inexplicable.

I'd love to hear which, if any of these ideas people would think most likely, or of any other possibilities I haven't considered. I'm a hard determinist so I would lean towards the first one because the second is a bit more 'soul-like'. But I can't discount the second because the nature of our 'qualia' seems to only be useful if experienced as it is i.e. colour as a labelling system allows for pattern recognition in a way that I can't see any neural network being capable of - and I am very much interested in neuroscience and understand how neural networks work. To me it seems that purely physical evolution could only get so far with the processing of light based sensory data before it came to a dead end, one that consciousness might be able to overcome if number 2 was right. But then you come to the problem of how consciousness could 'evolve' if it was not entirely dependent on the brain, but that's for another day

I realise my number two might come across as an argument for the soul but it is not meant that way. The way I envisage the mind is essentially virtual reality software running in the supercomputer that is the brain. Therefore in some weird way it is not necessarily that the processing in that program is happening on another plane but rather that it has been abstracted away from the raw data in much the same way as a high level programming language abstracts away from the pure machine code - in the end machine code runs but the high level, object orientated languages, give it an abstract and understandable meaning.

1.  I do't think that an explanation of mirror-of-process would make mind superfluous.  I don't see how, if it were a mirror, it could just as easily be there as not.  It would have to be there, if that was the process.  If it weren't there....we'd notice the difference, like we notice the difference(s) between a rock and a human being.

2.  What sort of processing are you thinking of, that couldn't be achieved by a nn?  We talking traditional problems...or something more eclectic?  I think that "qualia" has become the new "quantum"  .  

On what grounds are we to judge the likelihood of either, and what would we hope to draw from that? Just wondering, so I can try to answer more specifically.

(April 4, 2015 at 7:49 pm)Rhythm Wrote: 1.  I do't think that an explanation of mirror-of-process would make mind superfluous.  I don't see how, if it were a mirror, it could just as easily be there as not.  It would have to be there, if that was the process.  If it weren't there....we'd notice the difference, like we notice the difference(s) between a rock and a human being.

2.  What sort of processing are you thinking of, that couldn't be achieved by a nn?  We talking traditional problems...or something more eclectic?  I think that "qualia" has become the new "quantum"  .  

On what grounds are we to judge the likelihood of either, and what would we hope to draw from that?  Just wondering, so I can try to answer more specifically.

Okay, by mirror what I mean is every single thing that we experience in consciousness, and every single thing that we think we're doing in consciousness - choosing to think a thought, thinking that thought, identifying an object in our visual field etc - is actually accomplished in the neural networks alone and for reasons unknown it is also simultaneously 'presented' in the form of consciousness but take that consciousness away and our behaviour would be exactly the same, including with reports of conscious experience (I guess).

As for the neural processing it's complicated and a long time since I thought about it in detail. But the gist is that pattern recognition that makes use of the notions of space and colour would need both to be represented in neural form. Colour comes in in the essentially RGB format but what to do with it after that? One neuron per possible colour? I think unlikely. More likely is that it stays in its RGB form but then it is just an array of values that I can't see anything being able to detect the patterns we see in consciousness - for instance if you looked at the the binary code of a digitised painting you would not see the same patterns as looking at it as a painting.

As for the grounds of judgment I just meant a personal opinion. If I'm in the wrong forum please let me know. I didn't mean this so much as an exercise in logic but just in opinions and ideas.

Anyway, if you reply I won't be able to reply tonight as it's nearly 2am and I need some sleep. So, good night

Why would our behavior be exactly the same if we removed a contributing factor?  I would expect it to be altered to an extent commensurate with the alteration in circumstance. 

Quote:for instance if you looked at the the binary code of a digitised painting you would not see the same patterns as looking at it as a painting.

You don't speak or think binary....for a machine using binary this is exactly what it sees...the very same patterns, translated to a different language.  I may not see the painting, but the machine that speaks/thinks in binary does. In the same way.....I may not understand the beautiful lyrics of a song in swahili........meanwhile...those damned swahilis are getting teary eyed.

If mind is not more than brain function, then why awareness? Why should the complex cascade of electrochemical interactions be subjectively experienced?

Given that mind is supervenient on the brain, my question is this: is it possible for a supervenient entity to flip the tables, and become the causal driver rather than a passenger who's just along for the experiential ride? In other words, is the mind, while rooted in the brain, somehow also separate from it? I'd argue this must be the case, or there would be no reason for mind to exist. I'd choose something like your option #2.

I see consciousness as being a highly conserved property of complex life probably
present in forms as diverse as cuttlefish, cats, parrots and humans.

It's common because it's useful.
Self preservation becomes much easier if self is recognized to be preserved.
Self awareness is a complex pattern in the brain used for self preservation, no
more surprising a construction of functional molecules than a leg
is when used for locomotion.
Your disbelief in the capacity of a neural network to sustain consciousness is
suspect given the brain's evident ability to do so.  

For a plausible meso-scale theory of how the brain creates consciousness I recommend Antonio Damasio:
The Feeling of What Happens.

https://en.wikipedia.org/wiki/Damasio%27...sciousness
In it, human minds are constructed of:

1. Protoself: Where a representation of somatic internal status is maintained and continuously updated.
   
2. Core Consciousness: A second system which monitors this internal movie.  This system provides moment by moment awareness of self.
   
3. Extended Consciousness: A third system, involving long term storage, maintains continuity of self perception over time resulting in a personal history.
   

Even the least complex of these processes is significantly abstracted from the level of synaptic membrane permeability.  We'll have to trust to Moore's law a while before detailed connections between the lower and higher functions can be determined.  The BRAIN initiative is wildly ambitious like Bush's Mars program or the Human Genome Project.  We'll see if it is useless like the one or useful like the other.


This study is hampered by the extreme complexity of the systems and ethical concerns over experimentation on individuals with minds sophisticated enough to be interesting.  Damasios work is based on case studies of brain pathology and disciplines of neurophysiology, anatomy and histology.  Not so much on neural networks as studied by computer scientists. 

Mind over matter is retarded. It is basically a pagan concept.

It states that the mind can overcome anything, yet reality contradicts that.

What is mind? No matter!

What is matter? Never mind!

Or so they say...



@emjay

I don't agree with your dichotomy 1 vs. 2;
What is this consciousness which you are talking about? Can you give me a short definition?

I don't disbelieve in the capactiy of a neural network to sustain consciousness. I disbelieve that there is any function of human behavior that couldn't be replicated as well by a philosophical zombie as by an actually-sentient human being. If the universe were purely a mechanical one, then there would be nothing but things doing stuff-- no consciousness required. Therefore, I have to include that the possibility of sentience is intrinsically included in the makeup of the universe: i.e. that it's no less a part of reality than gravity.

(April 4, 2015 at 7:28 pm)emjay Wrote: As far as I can see there are two possibilities of how consciousness relates to the brain:

1. That it is the mirror image of processing in the physical neural networks of the brain and is therefore superfluous. It could just as easily be there as not because all the 'calculations' are performed by the underlying physical hardware.
2. It is more of a 'black box' that is an actual processing element in the brain and therefore 'interfaces' with the physical hardware in some way. In its other plane of existence, whatever that may be, it can perform processing that cannot be achieved by the neural networks alone. In this idea if (some of) the neural networks of the brain were essentially relegated to the roles of input and memory only, their data would be 'set' by processing in this other plane and a hypothetical examination of the brain would show changes in value seemingly inexplicable.

I'd love to hear which, if any of these ideas people would think most likely, or of any other possibilities I haven't considered. I'm a hard determinist so I would lean towards the first one because the second is a bit more 'soul-like'. But I can't discount the second because the nature of our 'qualia' seems to only be useful if experienced as it is i.e. colour as a labelling system allows for pattern recognition in a way that I can't see any neural network being capable of - and I am very much interested in neuroscience and understand how neural networks work. To me it seems that purely physical evolution could only get so far with the processing of light based sensory data before it came to a dead end, one that consciousness might be able to overcome if number 2 was right. But then you come to the problem of how consciousness could 'evolve' if it was not entirely dependent on the brain, but that's for another day

I realise my number two might come across as an argument for the soul but it is not meant that way. The way I envisage the mind is essentially virtual reality software running in the supercomputer that is the brain. Therefore in some weird way it is not necessarily that the processing in that program is happening on another plane but rather that it has been abstracted away from the raw data in much the same way as a high level programming language abstracts away from the pure machine code - in the end machine code runs but  the high level, object orientated languages, give it an abstract and understandable meaning.

I would first make a distinction between some of these terms, something like the following:
Sensation - a supervening principle of neural networks; it is reducible to an inevitable functionality of certain physical objects when they interact in a specific manner. How an environment feels to, say, a worm, may be different than it does to a slug, both of which are in a class above say, a single-celled organism, which lacks any sensory input, but are obviously lower than something like a cat or a dog. There are probably slight differences between every animal, the more obvious separation, though, maintained between species.
Consciousness - an evolution of the processes described in the preceding definition, where neural networks have reached the point of development to be contained in their own central region, in a distinct organ, and often (though not necessarily) receiving the additional input of such features as vision and sound. 
Mind - A further development, per the unspecified principle(s) of organized material compositions, where conscious sensations are transposed to symbolic representations, enabling us to grasp our surroundings in an intelligible manner, e.g. the rational animal, man.
It would seem to me that consciousness and mind depend on the development of the brain, but that sensation precedes both. Colors enter the foray at the level of consciousness, whereas something like the following enters at the level of mind: 
Given my inability to formulate a coherent understanding of what abstract objects, such as the principle the above statement describes, are in relation to the material world, I might at this point take the route that only mind discovers reality (in part) as it truly is, and that is something both material and abstract (call it neutral monism).