Quantum consciousness...

[The Grand Nudger]

While Robert Epstein, the psychologist who wrote that article I posted didn't always make his argument particularly well, I think he is on the right track in pointing out the limitations of describing the brain as a machine. You need to see it in a historical context. Since computers became useful, scientists have been trying to describe the brain as a machine, but decreasingly so the more we understand how the brain actually works. I have read some truly painful papers from the 60's in this regard.

OFC, and I loved that part of the article actually.  While, ultimately, our historical explanations for consciousness have as yet turned out to be failures - they were good steps into the unknown.  I think that you take an uneccessarily dim view of the mechanical.  So long as we are attempting to explain the mechanism of consciousness, we are explaining the brain as a machine.  It may be a machine unlike any other, but it's still a machine. I;ve seen terrible computer analoigies as well, but I;ve never seen a terrible comp theory.  

Consider this, even the theory of "humors" was a step in thew right direction.  We observed complex behaviors being accomplished by pnuematic devices, and it dawned on us that - rather than some ephemeral spirit, perhaps their were physical interactions within our bodies that produced our own observed behaviors.  For all that this truly painful theory (at least, to our eyes) got wrong...it got that much right - and this was a revolutionary concept of consciousness in and of itself.

The brain is a self organising system. All evidence points to that and there is no reason to suggest that it isn't. There is absolutely no evidence of a non-corporeal soul directing the brain, and even if there was then that soul and the brain should together be considered a self organising system. Dendritic trees self organise. Neurons self-organise (e.g. habituation). Emotions can be understood in terms of self organisation. There is no reason to think that consciousness is any different.

-and a comp theory would be the -how- of that self organization as it applied to consciousness.

The computational framework does not adequately explain self-organising systems. Natural scientists such as chemists, geologists and physicists do not describe naturally occurring self organising systems in terms of computation. A geologist will not describe how plate-tectonics push up mountains in terms of information theory. It is just not a useful concept for describing such things. Natural scientists understand self organising systems in terms of pressure, minimising the flow of free energy, settling into stable states, minimising entropy locally but maximising it globally and in terms of thermodynamic gradients.

Comp sci doesn't adequately explain the properties of the materials that form the computational architecture either.  For a moment, imagine that one of the many comp theories had merit - that it was meaningfully accurate - it would still require a unifying theory of biology to explain how it was that the material of the computational architecture, the brain, came to be in a state capable of performing the stated actions.  

Information theory was derived from the concept of entropy. It is a higher level of abstraction. What information processing is required for a snow flake to form? The brain is also a naturally occurring self organising system so why do we make a special case and describe that in terms of information flow and computation? Possibly because it is so complex and an information theoretic approach allows us to abstract over the details more. For example, you wouldn't describe a steam engine in terms of information theory but if you had millions of little steam engines all working together in a complex way then I can imagine how it would be useful. It is still an open question as to what stage an information theoretic / computational approach becomes useful and why.

I wouldn't describe a steam engine in that manner..but I could design, build, and describe a steam computer in that manner.  

Describing the brain in terms of computation may allow us to reason about what it is doing, but not why. For that you need to recognise what it fundamentally is and why it came about.

It's more the how than the what.  The "what" is what compels us to refer to comp theory.  It -seems- as if the brain is processing information.  From the inputs of our sensory system to it's own internal memories - and the product of this processing is not only reliable - but somehow standardized between individuals.  This l;eads to vast and compelling similarities of behavior and even our ability to understand and predict each others response and "interior experience".  It also leads to our ability to store and transfer those interal experiences - in a variety of media, to other individuals in which this "x" is substantiated and even to machines in which we don't think this "x" is substantiated - albeit in a very limited way.   The brain has likely been as you say it;s been for a very long time, long before it ever developed into something that we would recognize as a conscious system. Long story short, it's productive and systematic.  

What accounts for the difference between the simplest brains and our own - unspecified complexity in a self organizing system, or specific complexity in the form of computational ability?  This is thew question posed and an answer tentatively given...with regards to comp theories.

Khemikal, you misunderstood what I was saying about temporal sequence learning. I was not explaining consciousness in terms of temporal sequence learning (TSL). I was using TSL as an example of intelligence that can be seen to aid self organisation. My point was that consciousness should also be seen as a way of aiding self organisation. It is a product of the brain and that is what the brain does. The development of consciousness is after all part of the same arrow of time where complexity has increased since the Big Bang. I am not saying that unspecified complexity complexity explains consciousness, I am saying that the increase in complexity over time is the result of naturally occurring self organisation which happens because of the laws of thermodynamics. The development of consciousness is another step in this process that has been occurring since the Big Bang and which has continued on to give us societies and economies. Consciousness allows us to act more intelligently.

Agreed, however, there is no comp theory that sets itself in contradiction to any of the above.  

I gave it a signal that told it how well it was doing. I did not specify what to do with that signal, but the strength of that signal could only be changed by temporal sequence learning.

What is "it"?  

I don't know. I didn't tell it what to do. It self organised.

Actually I have since visualised it in action and now have a vague idea as to what it is doing, and it surprised me, but I don't want to say because it's the basis of my next paper. It's also somewhat complex to explain.

I'd love to read it when you get around to it.  

There is no algorithm specifying how it should perform temporal sequence learning. What I did was create a system that would settle into a stable state. The trick was in creating the right components. I have no doubt that I could have used my neural networks to do it instead, and if so then I could recreate it my garden using a system of pipes and water cisterns. Such a system would not be adequately described using an algorithm but by the changes in water pressure within it.

The stable state, if the stable state is what allows it to reliably perform some action x in future, would be a machine implemented algorithm, the structure of the machine itself would be a form of memory.  Stable states are a requirement of discrete and reliable memory.  Overshadowing this entire business is that your net model is most definitely playing itself out on a computational architecture.

Neural net theory is a computational theory.....the connectionist comp theory.........there is a current of thought that asserts that, as far as the human brain and human behavior is concerned, connectionist theorists are indispensible - and likely account at least in some way for what -any- brain is doing...even ones that do not seem to produce a recognizable consciousness. Theyre much better than formal comp theories at explaining motor response and perceptual frameworks - but they fail to explain or account for productive, systematic, and standardized manipulation of precepts. This is why many consider the next "productive" comp hypothesis to be some synthesis between nueral biology, connectionist comp - and some as yet undetermined symbol manipulation theory. Maybe one that's already been floated, maybe not. Consider, for example, the seamless amalgam of a connectionist mechanism applied to dennets multiple drafts model - a model which emphatically -denies- what many of us take consciousness to be. I don't propose this as the answer, only to highlight how different types of comp theory compliment each other.

Or maybe it;s all as wrong as "humors" - and if that were the case, if the comp framework is inadequate, so too are NN's as a representative subset of that class. As, in the compter analogy, giant bundles of memory substantiating machine implemented algorithms.

---

I'm interested in this sentence, because I think it lies at the heart of one's view about consciousness.  It seems to me that consciousness is binary-- it's like a light switch that's either on or off.  The particular ideas, sensations, and so on that one might experience seem to me to exist IN consciousness, or perhaps to be objects OF consciousness.  I wouldn't say that something is more conscious, but rather that it is more mentally active, and I wouldn't equate mental activity with consciousness-- at least, not exactly as synonymous.

On the other hand, what would a conscious agent be like if there was NO perception or sensation?  Would it still be conscious, or is that simply an oxymoron?  I do have some interest in meditation-- certainly those guys, whatever you think of their spiritual or religious beliefs, are really walking close to that line between consciousness and a lack of it.

Human beings subjected to extreme sensory deprivation report, in addition to many other things, a loss of self or conscious experience.  OFC, they also report all manner of conscious experience.  It;s almost as if the brain plays itself out for as long as it can (sometimes terrifyingly) and then, at some point, deprived of stimulus it throws in the towel. In that context, it doesn;t seem binary, more a gradual loss of fidelity ultimately terminating in an interuption. Consider, again, what is reported upon leaving a dep chamber. A sudden "reawakening" and rush of experience. At least in this sense, I don't know if consciousness can be accurately described as being on or off. Is it on or off when you sleep? At what point, in sleep - as another example, do you go from dreaming to "blank"?

On an esoteric note, this degradation, loss, and resumption of conscious experience is precisely the aim of a great many ritual forms....down to simple reps like a traditional lakota sweat lodge...but it's also a part of recreational drug use(lol).