Before joining the conversation, please read and accept this Invitation to a Conversation.

This is a brief discussion that touches on Turing machines, neural networks, universal computation, system theory, system matrix notation, cosmic consciousness, individual consciousness, systemic evolution and holistic science.

I previously mentioned the mathematics and its computational implementation that arose from my metaphysical research in the article IT Revolution. Here I'll discuss how this fits in with system theory, consciousness, metaphysics and the evolution of systems from particles to civilisation, but first I'll begin by saying a little more about the mathematical / computational process by describing a simple way of thinking about it in terms of Turing machines and neural networks - it is computationally equivalent to a massively parallel network of neurons but we can work towards it by thinking about Turing machines. Let me explain in 4 steps:

Step 1: Simple Turing Machine

Consider a simple Turing machine T1 that has a tape of passive elements that can store data that can be written or read. There is also an active head that can step along the tape and read the data from the passive elements. The head has an internal state and it reads the input data from one element and in combination with its internal state it maps the input state into an output state that it writes to the tape and may also change internal state. Different input/output mappings and state transition mappings result in different algorithms. This simple scheme can give rise to a universal computational process that can implement any computable algorithm where a 2:5 Turing machine is the simplest known universal Turing machine with 2 internal states and 5 distinct data values.

Step 2: Complex Turing Machine

Now consider a Turing machine T2 that doesn't step along the tape and read the passive elements one at a time, but instead it can read in the entire tape as a single input. But it can only write to a single element. This is equivalent to a Turing machine with a single passive element that has a large number of data values. E.g. consider a tape with 8 binary elements, instead of reading single bits one at a time it can read a single 8 bit value so instead of 8 elements with 2 values it is one element with 2^8=256 values. T2 can read in this one value and in combination with its internal state it maps this to a single output value within the range 0-255 in which only one of the bits is allowed to change and the rest remain unchanged.

Step 3: Single Neuron

Now consider a Turing machine T3 where its internal state isn't separate from the tape but is instead stored in one of the data elements so one element, say the first, represents the internal state of the head and the rest represents the input data. We'll call the first element here an active element because it is associated with the active head. So T3 can read in the entire tape which includes both its internal state and the input data. This is then transformed into an output state but the only element that it can change is its active element that represents its internal state and the remaining data remains unchanged. This is purely an observer of the passive elements, it cannot change them, it can only observe the passive data and set its internal state accordingly. On its own this is not a very universal computational process but is a simple model of a neuron. If this was all there was the passive data would never change and the observer would repeatedly observe the same input data and only its internal state could change. Hence a single neuron is not very useful.

Step 4: Neural Network

Now consider the case of a neural network where every tape element is an active element with an associated head. Each head reads in the entire tape and the n'th head treats the n'th element as its internal state and the other elements as its input data. Each head reads in the entire data that includes the input data and its internal state, then it changes state and writes this to its active element leaving the rest of the tape unchanged. In this case each element of the tape is the internal state of an observer or neuron and each neuron is observing the entire tape. Because each element is the internal state of a neuron and each neuron can change its internal state the tape elements can continue to change so the input data that each neuron is observing keeps changing. Because of this the internal state of each neuron keeps changing and the changes lead to further changes and so on.

Systems

Every neuron can potentially interact with every other neuron so there is potentially no distance between them but they don't need to be this interconnected. For example, a particular neuron may only pay attention to particular neurons or it may give more weight to some than others, and there may be only a particular group of other neurons that pay attention to its state. This creates a complex but localised network of interactions or information channels that binds certain neurons into functional groups. Some paths may open out to other neurons for input and output and others may form closed loops. These functional groups act as systems and as the interactions evolve the systems integrate and disintegrate as sub-systems form into super-systems and super-systems decay back into their sub-systems. Thus a neural network is a "general system" simulator or a systemic universal computational process.

A Computational Mind

This computational process effectively creates a closed massively parallel neural network where the state of the tape represents the internal state of every neuron in that network or the holistic state of the network. Different initial internal states, neuronal interconnection patterns and state transition mappings result in different computational processes. The state of the neural network can be called a "state of mind" and each state of mind flows into other states of mind as the configuration evolves. This scheme can easily be implemented using extended matrix algebra that liberates it from being constrained to linear systems. In the extended scheme called system matrix notation (SMN) any linear or non-linear computational process can be implemented. Refer to Finite Discrete Information Systems to see how this is done.

Cosmic Consciousness

The network is completely closed, where the network state is the state of the virtual universe, which is like a dream state within the mind of a cosmic consciousness. The neurons implement primitive systems, these are indivisible systems that interact and can integrate to form compound systems that are functional groups of neurons. These can further interact and integrate to form higher levels of systems and thus the virtual universe takes on a system theoretic structure with systems within systems within systems. The entire neural network is the largest functional group of neurons and it comprises the system that can be called the universe. Within this virtual universe every system can potentially interact with every other system so there is no intrinsic distance between them but as they form into functional groups and certain signals need to travel through a network of systems in order to pass between particular systems the concept of separation or distance arises. A regular metric can create any kind of dimensional space (e.g. 3D space) and non-regular interaction patterns can create other kinds of fractional dimensional or non-dimensional spaces (e.g. the internet).

Individual Consciousness

In the case of individual virtual systems these are functional groups within the universal network so they are not entirely closed. Each virtual system is a sub-network within the universal network so each system is a microcosm of the cosmos. The main difference is that they are open systems that have an internal network that opens onto the wider network and they interact with other sub-networks within the universal network. Whilst the cosmic network is closed and there is only an inner space, the virtual systems experience having an inner and an outer space. Within these sub-networks some neurons can be observing inputs that open outward from the sub-network, which then stimulate the sub-network to respond and evolve according to its nature. Some elements or neuronal states can also be observed by other networks and can thus be used as outputs to influence external systems. The inputs are sensory inputs into a computational mind and the outputs are actions driven by the computational mind. In this way the mind can evolve or contemplate within its own internal space or it can experience sensory inputs and respond with output 'actions'.

Systems can be primarily contemplative with a large portion of their sub-network devoted to internal processing or they can be primarily reflexive with a large portion of their sub-network devoted to translating from input to output. If they are primarily contemplative they can process their inputs deeply and form complex internal spaces of awareness, imagination, knowledge and so on. If they are primarily reflexive they respond in simplistic ways to their inputs with little internal reflection.

Perception and Reality

When systems perceive through their senses they are embedded in the information stream so the perceived systems appear to be objects in space where the objects are tightly interacting systems or functional groups of neurons and the spatial distance between them arises from the interaction separation of the systems where the information must flow through a network of intervening systems in order to be conveyed. But underlying this every system is directly connected to every other system and there is no distance between anything. At the level of the information flow things don't appear as objects in space, instead there is a vast flux of information streaming in every direction and interconnecting everything at every level.

Complex Systems

As the virtual system evolves from simple systems toward more complex systems there are a large number of low-level systems that are highly reflexive to the point of being automatons. These are just cognitive-feed-through components that can are connected together and programmed to elicit standardised reflexive behaviours. However the higher level systems are fewer in number and they are more contemplative and able to process and cognise information more deeply and are thus able to engage in more complex and variable behaviours.

In the virtual universe the simplest systems are the most reflexive and the higher level systems are more contemplative, but with each system level they also tend from reflexive to contemplative. As they reach a level of internal complexity they breach a threshold where they are able to engage in more complex communication and integration and at this point they integrate to form a higher systemic level of systems. This is called a meta-system transition.

Systemic Evolution

In the following example I will use many common labels such as particle, membrane, cell, etc but remember that these are just perceptual analogies for what are actually virtual systemic structures or dynamic functional groups of neurons within the cosmic reality generative neural network. They are dream objects within the cosmic consciousness and not 'material' objects in space. But when perceived by systems through their senses and interpreted using a materialistic paradigm they are thought to be material objects in space.

The lowest level systems such as particles have almost zero contemplative capacity and are almost entirely reflexive but these integrate to form atoms and molecules. At the level of bio-molecular structures the systems are complex enough to be able to integrate to form an entirely new level of systems called cells. These prokaryotes are simple membranes with a DNA-RNA-protein cycle and they persisted for billions of years as a single cellular ecosystem. Then the eukaryotes formed which are an outer membrane with inner sub-membranes allowing for a more complex internal processing so they are thus more contemplative. Very quickly they produced an entirely new level of systems called multi-cellular organisms. These formed ecosystems where the lower level systems are bacteria, insects and plants, which support more complex systems such as tigers, elephants, dolphins and humans that are much fewer in number and have much greater contemplative capacity and less reflexive behaviour.

Humans have developed far greater contemplative capacity than other animals and we have thereby integrated to form an entirely new level of systems called organisations, or tribes, societies, corporations, nations and civilisations. And the process of systemic creation continues, but all the systems are still just virtual systems or dream systems within the cosmic consciousness. There is still just the same neural network but its configuration or network state is evolving.

Other related discussions here at NCN:

The Gaian-Ego Hypothesis
The Scientific Case Against Materialism
IT Revolution
Systems Analysis of Economic Social Engineering
The Mystic Meaning of Original Sin

More articles follow on from these going into other related aspects (see the bottom of each page). All these issues are connected because they are all configurations and behaviours within the one cosmic network. We perceive things as separate and form separate discourses for them such as physics, psychology, spirituality, politics, economics and so on, but everything arises from the one unified source and everything can be conceived of as virtual systems interacting within a cosmic consciousness. This understanding is the basis of holistic science.

There is also more detailed information on my website System Theoretic Metaphysics of Reality.

Best wishes : )
John Ringland

Before joining the conversation, please read and accept this Invitation to a Conversation.