User:Graeme E. Smith/GreySmith Virtual Architecture Theory

 Virtual Architecture Theory 

In this theory, an Architecture is a structure, within which machines work. A Virtual Architecture is a structure in which Virtual Machines work. A Virtual Machine is a mechanism where, a machine of a particular architecture, is run on a machine of the same or a different architecture as if it were a different entity. The Virtual Architecture of Mind Theory, states that the mind is made up of a number of machines in a structure, and that some of these machines are Virtual Machines in that they run on machines of a completely different architecture. In this theory, a biological element whose function is fully known can be replaced by a machine, and therefore operates as if it were a machine, and is so classified.

Consciousness, if it even exists, represents a specific architecture of mind, that has certain characteristics which are knowable. It is thought that the particular architecture of mind needed to achieve primary consciousness, is the result of vertebrate evolution, and is probably limited to the Birds and Mammals. It is thought that the particular architecture of mind needed to achieve Higher Order Consciousness, is limited to the Hominid Species, of which there is only one species currently known to exist, that of Homo Sapiens.

Where a biological element is fully known it can be replaced with a functional equivalent, However where a biological element is still incompletely understood, it can be better replaced with a simulation that presents as many currently known elements as is possible. This theory recognizes that it comes at a time of great change in the understanding of how the brain works, and that thus, it must be at least partially a simulation, rather than a Functional Model of the brain. This theory is therefore intentionally left incomplete with the intention of upgrading the simulations as new versions become possible.

Because this Theory is part simulation and part Model, parts of it fall within the purview of the scientific disciplines that say that consciousness must be simulated to be achieved, and part of it fits within the purview of those that say that consciousness must be modeled to be understood. It is therefore a Hybrid Theory, with portions that fit within the Phenomenal or simulation disciplines and portions that fit within the functional or Modeling disciplines. It should be noted that the portions that are phenomenal in nature are those elements that are the hardest to model, and the portions that are functional in nature are those parts that are the best understood.

If there are two types of Consciousness Architecture, we can't at this time tell them apart, since there is no known difference in the mechanisms of mind, between mammals and homo sapiens, partially since homo sapiens is in fact a mammal. So we must model only the general Primary Consciousness or mammalian architecture, and hope once we have learned enough about it, that our model will show us where the differences are that create the sophistication of the homo sapiens Higher order Consciousness. The primary differences seem to be thought, and Language, and these may turn out to be trivial once we understand the rest of the architecture.

At this point I usually joke that I am trying to capture the consciousness of a Rabbit in my Theory.

Since both the Phenomenalist and Functionalist Schools claim to have developed consciousness, this theory might be seen as un-needed, however it offers something that neither of the other schools do, a basic insight into the nature of consciousness that lets us define it more closely than was heretofore possible. Of course to get there it breaks some of the rules of both schools of thought, and therefore draws itself into question, but it opens up a new vista, a new paradigm of what is happening in the brain, that while it can't be proven to be true at this particular point in time, opens up some questions that I think are important to ask.

The basic architecture of this theory is a mind architecture that incorporates a memory architecture Virtual Architecture Layer with a Mind Architecture Virtual Machine that runs on top of it. It is interesting to note that while the Memory Virtual Architecture is unique, and relies very heavily on phenomenalist principles, the Mind Architecture Virtual Machine that runs on top of it, is very much a Functionalist Mechanism, if only because our simulations do not run far enough to adequately simulate the mind architecture. It is a matter of science, simulation technology lags our attempts to model the functions of the brain, so even if we wanted a full phenomenalist simulation of the brain we couldn't yet achieve it.

My book, How the Memory Might Work (HMemMW) explains the basic Memory Virtual Architecture and how specific organs in the brain contribute to form a memory that involves:


 * An Implicit Memory Layer
 * An Explicit Memory Layer
 * A Declarative Memory Layer
 * A Meta-Index Layer, and
 * A Consolidation Layer

Each of these layers can be seen to be a virtual machine running on the previous layer. The Implicit Memory Layer running on the essential phenomenal Neural network layer. It may not be the only interpretation of the brain, nor even the best interpretation of the brain that is possible but studying the interfaces between the layers offers us a unique perspective on how the brain might work. Of critical importance is the interface between the implicit and explicit Memory layers, as that defines the bottleneck. It is the nature of the Implicit memory as a phenomenal system that requires the bottleneck, and influences the nature of the attention systems used in the brain. Incidentally, this model deals well with the Qualia problem that phenomenalists push as a test of the Qualities of conscious experience.

My second book How the Mind might work (HMindMW) explains the basic Mind Virtual Machine that runs on top of the Memory Virtual Architecture. Unfortunately neither of these books are yet published, and so it would be premature to expand on the theory. However one of the characteristics of my Mind Model, is its dependency on something I call an AM Language short form for Adaptive Macro Language which uses advanced compilation mechanisms to adapt a basic Macro language in a way that overcomes some of the shortcomings of past macro languages such as the Forth Language.

Consciousness, I explain, in HMindMW is an architectural requirement in order to implement a self-programming system using an AM Language. However the Model would not be as complete without the Memory Virtual Architecture that supports it, since it would not implement qualia properly.

It is because I need both phenomenal and Functional components to complete my model of consciousness that I question if either alone could do so. --Graeme E. Smith 17:30, 11 February 2009 (UTC)