11 July, 2012

Turtles all the way side ways...Godel and Turing point the way.

In a response to a friend on Facebook an interesting discussion on recent results that decidability of paths for particle transmission at the quantum scale is not assured similar to Turing's halting result in computer science and Godel's incompleteness in logic. This article was the original source of the discussion. Below I respond to comments made by my friend.

"Making harder the conundrum is that Godel's Incompleteness and Turing's halting only apply to countable sets, not continua."

-- Precisely, but the riddle being that an infinitely countable set is included in that definition...and it's possible to have infinite subsets of such sets! So discretization (of anything) seems to be fundamental to continuity...this truth of mathematics which stands apart from our present understanding of reality may point us in the direction of what reality really is about.

If the pattern continues to apply it may provide a way to test validity of multiverse theories. As you may know, there are several interpretations of the results of quantum mechanical theory and general relativity that seem to indicate the flowering of either event histories, or the multiplicity of entire Universes...but the question is, are these histories or Universes real? Which is the truth? Mathematically, the application of sum over histories in Feynman's formulation of QED enabled the solving of previously intractable problems. The solution involved the idea of wave function collapse (sum over) over an *infinite* set of possible histories for a given event..each with variable probability of occurrence.

Some interpreted wave function collapse as meaning the spontaneous creation of new realities to evolve the state of the system, others see the multiverse as a coeventful (I can't say "temporal" since measurement of time may be different between Universes, so I use "event" to encapsulate the relative difference in time measurement that would attend different Universes...according to GR, having a time dimension at all is not even necessary) set of Universes that may or may not exist *now* but did at some time in the past or will exist at some time in the future.

I attended a Physics symposium in 2010, where I was fortunate to meet and hear speak David Gross, Micio Kaku and Alan Guth. During a spirited discussion Gross and Guth battled over whether or not they felt the idea of multiverse made sense. Guth imagined there could be a multiverse of universes and posited possible travel between them...Gross called it nonsense saying that the mathematical existence of multiple Universes does not guarantee coeventful transit capability (which I agree with).

I am reminded of how we now understand energy in the modern sense that is different from how it was described classically when one studied 19th century energy based on newtonian physics. Then, we were told that "energy can't be created or destroyed" in the famous second law, it turns out quantum allows it to be created and destroyed so long as over time those events even out. *There is a sum over the histories* of not just particle evolutions but particle existence...if that's the case it means a) all of this truly can be the result of the spontaneous eruption of something from nothing. and b) that in order for balance to be restored this "something" will and must go back to a "nothing" state. This  though sounds a lot like what was supposed earlier about a coeventful multiverse. What if it is enough that the ensemble (infinite) of Universes goes through all possibilities for each contained particle event... *eventually* rather than that they all occur coeventfully. It would solve the multiverse problem by applying Feynman's idea of sum over histories to the entire multiverse itself. It's actually beautiful....and fractal at core...and also...in it's universal absolute...very binary, like the divisions between the integers upon which both Godels incompleteness and Turing's computation halting problem are based....coincidence?? I don't think so.

""What if" recursion (iow fractal nature) is so Total, it's wholeclothe with dimensionality?"

I considered this but it would seem from what we know thus far that reality is not "turtles all the way down" as we look deeper into the fabric we see discretization not continuity. We go down to the atom and see it begin in energy transitions of electrons in their "orbits"....allowed and forbidden with no middle ground, we see the emergence of macroscopically absent dimensions of discretized energy transduction...like "spin" (intrinsic angular momentum) ....we go down further still and see discretization in the family of quarks that make up the particles of the atom. Deeper still and we reach the limit of the planck length...below which space itself is a roiling soup of undulations that emerge and eliminate particles.  For the recursion idea to hold *within our universe* there would have to be a continuous ability to probe deeper...so far all our physics...theoretically falsified or unfalsifiable (string theories) point to a bottom. A possible solution though lays in if additional universes *pinch off* dimensionally from parent Universes such that all their dimensions are orthogonal to all ours...this would then make them independent and maintain the fractal symmetry but instead of going "down" it goes sideways...as it were. This idea is one I am partial to (and one that received media last year when a physicist proposed it)...and the fact that GR predicts objects that are forever hidden from the Universe that created them (black holes...namely the singularity within) it just may be the "sideways" set of dimensions in which new Universes (and their physics and ability to create black holes) exist....and thus maintain "turtles all the way sideways". The continuity is preserved through the fractal recursion...the discretization is preserved through the independence of the new Universes birthed...riddle solved.







08 July, 2012

Objections to a non physical conclusion for consciousness.

In a paper recently published by Mark Muhlestein in the journal Cognitive Computation the following conclusion is made:

"In this computational framework, the distinction between a computation and the recording of a computation can be blurred arbitrarily, yet the physical implementation of the computation itself is unchanged. From this, we conclude that a purely computational account of consciousness is unsatisfactory."

This conclusion I agree with but for a few reasons not directly addressed in the paper which I elaborate on below.

On the objections, I'd say the ones that are most against this conclusion (that either computation is not all there is or consciousness is not what we think it is, roughly) are objections 3 and 4.

The clarion call that was ringing in my head as I read was the condition that a random number generator was being used. Conscious states are intimately dynamic systems that do not have deterministic response as asserted in the thought experiment because unlike the real woody, the woody emulation or the projection, real woody is subject to the vagaries of internal machinations that do not emerge from random desires or even external ones. These are the autonomic drivers that internally modulate I assert, internal characteristics of cognitive dynamism at the layer of the neocortex from the deeper layers of cognition in subtle and non repeatable ways.

Thus the failing of this paper, a devastating one lies under the assumption that determinism in emerged experiences is even possible. Repeated stimuli may emerge "similar" reported conscious states would never report identical experience states as real cognitive states are ultimately driven by physical sensory modulators that run continuously and are distributed in more places than just the computational core in the mind. How would the woody emulation for example 'simulate' the time dependent process of slowly becoming hungry over time and how that varies cognitive state? or of slowly feeling hot? or of itching at the fibers of wool in the sweater one is wearing while trying to smell that Rose? In a simulation such inputs are assumed to be of in a fixed (and undetermined) input state but I posit that is precisely why any subsequent question regarding derived experience can not be said to be deterministic and can not be said to be identical given "identical inputs". Not only is there feedback between the processing elements in the emulation and other processing elements (that naively in this thought experiment do the "computation")  but there is also feedback between *internal sensory states* modulated by external sensory states that are continuously varying. 

The *illusion* or non physical appearance of consciousness comes from this fundamental disconnect in how it is being characterized in this thought experiment as being purely a process of computation once inputs are gathered, in Bishop's conception and in Maudlin's conceptions.

I assert that Inputs are continuously gathered and modulate the "computation" if you don't simulate that in your simulation what you emerge will indeed be a shadow of consciousness.

Moreover, near the end when he states:

"Evolution has not prepared us for interactions with an entity which exhibits conscious behavior but which is not in fact conscious; we would find it tempting, perhaps overwhelmingly so, to grant full rights and responsibilities to an entity which can pass every imaginable test of sentience. But if Woody says, “Come on in! The water’s fine!” and suggests that you submit to a destructive brain scan in order to upload to a computational substrate, you would be wise to do so only if the nature of consciousness is clearly understood, and if you have included as part of your upload package any hardware (or wetware) necessary for conscious experience."

I agree that caution should be had here for more fundamental reasons. I assert that such a process of copying the minutia of ones computational substrate is insufficient to copy ones consciousness substrate which includes elements distributed through out the physical body. Again, the cognitive system emerges from intimate sensory and physical contact and modulation from both external and internal states which are dynamic...any computations are continuously converged to and are necessarily unique in approach...even if  the input signals are identical. The convergence to an indicted state of experience will always be slightly different, consciousness can still be intimately tied to those insensory , exsensory and cognitive processing interactions, still be physical. Those that are all gung ho about being "uploaded" need to realize that what will be done strictly (if possible) is a copy of a cognitive brain state (absent the inputs of physical modulation!) and then that will have a hell of an *awakening* once it boots in the artificial substrate. Furthermore, what boots up won't at all be the person copied but a simulacrum....the original persons consciousness would end with the destructive copy process...and the now pathological copy would go on in terror in the copied substrate. Even if the entire physical feedback loops of the original individual could also be copied it would still again be a copy and NOT the original agent. 

I can imagine only one scenario where he original consciousness can slide to a new non biological substrate and that is if the original body is piece by piece replaced by non biological elements across it's entire physical construction such that a new consciousness is not created but rather the existing one is moved (but in place) to a new substrate which includes both cognitive replacement and propriosensory and somatosensory replacement.

Now all that said, do I agree with the initial conclusion? Yes, and No. Yes, I agree that cognitive computational processes (with processing units resident only in the brain that are fed specific inputs) are not sufficient to replicate consciousness. However, No I do not accept the often ready fall back that because emerged consciousness is not entirely resident in the brain (but rather in the brain body physical unit) that it must come from some where external to the physical body entirely as some would like to fall to, or that it involved spooky quantum interaction among the cognitive elements that can not be simulated in an emulation.