Homogeneous & Heterogeneous

Homogeneity & Heterogeneity

Let's dive in.

Introduction

This planet that we live on began as a blank slate, one where the geology gave rise to relatively simple processes, processes where the entropy was increasing, due to the radiation of heat from that geology in the Hadean, the first and oldest eon of Earth’s history, lasting from the planet’s formation about 4.6 billion years ago until 4 billion years ago . The energy from that radiation was not being utilised by the planet, and further to this, there was no life to absorbed the radiation from Sol, our sun. When as the planet cooled sufficiently, life first arose, simple at its inception, but gradually developing into more complex cellular systems. This complexity resulted in multicellular life, which includes you, me, and all of the environment. Most of life extracts its needs from the energy of the sun; either directly (plants) or indirectly (animals). Life has populated the whole planet, creating diversity, where all life, forms an immensely complex system, one where each organism plays its role in. This has led to a far more heterogeneous system, one that is an intelligence in itself. Just as diverse neurons interact in intelligent ways, so does all of life.

You are constructed from and think through a highly heterogeneous set of systems. Each neuron is unique, interacting with populations of unique individuals. But your brain is not the centre of your consciousness, consciousness is dissipated throughout your body, your senses bring information to the brain, and those sensations are a part of your consciousness. Further to this, your intestine is populated by neurons that interact with the brain, it’s almost like you have a second brain, one that impacts upon your conscious experience. Within your intestine is a diverse population of microorganisms, called the microbiome, including bacteria, viruses, and fungi. These populations are an ecosystem in themselves. They respond to, and influence what you eat; there are many different types that increase and decrease dependent upon the environmental conditions (what you’ve eaten and what other microorganisms are present). Different populations exist at different locations, depending upon what they consume to survive, some surviving upon the waste of others, the waste created through processing what you have ingested. As you can see these systems are highly varied, just as the population of the neurons within your brain are.

Your brain is subject to the environment that you are embedded within, it is programmed by that environment. Your intelligence is due to understanding the environment correctly, and acting upon those understandings in a correct manner, a manner that allows for your to survival, through strategies that are in line with the rules of the environment and reality. To understand that environment, intelligences need neural circuits that are tuned to the components and processes of that environment, whatever that environment may be. Your neural circuits evolve with the environment, each neurons connections being strengthened or pruned, depending upon the information that is being fed to those neurons, though senses or internal processes such as thoughts, or the microbiome. That tuning creates neural circuits, those neural circuits can be replayed, creating the thought necessary for a specific understanding. As you may realise, we need highly heterogeneous mixture of circuits to process each of those objects, objects of the environment. When you change environment, your neurons will need to tune to the new environment as you learn about it, to become in tune with it.

This is the reason that Heterogeneity is so important a foundational principal of intelligence. Life is heterogeneity upon heterogeneity, complexity upon complexity.

Homogeneous

Homogeneous from the Beginning

As was discussed on the Systems page, the universe is progressing to a more chaotic and disorganised state, or what is known as increasing in entropy.

But the early universe was incredibly homogeneous, there were none of the atoms that life, and you and me are made from, to function as we do. Yes there were the hydrogen atoms in the water that is a part of you, but these cannot interact in the complex ways that are necessary for intelligence to arise within a system. There were none of the carbon, nitrogen, oxygen, phosphorus and the many other types of atom that your DNA and cellular systems are constructed from, these all came much later, as a product of the first and successive generations of stars that populate the universe.

Due to these facts, it is highly improbable that intelligent organisms existed very early in the evolution of the Universe. 

Intelligence requires understandings of complex systems, to be able to navigate those complex systems. Complex systems/environments did not exist in the early universe, so there was no reason for an intelligence to have to navigate such an environment, and therefore there would be no need for neural circuits to arise and be programmed by such an environment. 

As you probably realise, this is a chicken and egg problem. Which came first, the complex brain or the complex environment?

Could the universe think for itself? This is a question that is commonly asked by those of philosophical and/or spiritual persuasions.

Whether this is true is debatable. What we decide to be ‘true’ is determined by our subjectivity and how we experience that subjectivity when we discuss the question with others, or internalise it within our own minds. But even then, believing one way or another is still due to the interaction of neurons and neural circuits. 

On the mindful page we discuss emptying the mind, and observing thought and experience as it manifests in consciousness. After practicing a while, one is able to empty the mind of thought, but there is still this residual thought conducting the observation. Due to these findings, one may realise that empty thoughtless space is almost like that early universe. Thought exists for a moment, and then is gone, it is not dwelled upon. Is there an ‘observer’ to that original universal void, a process that can observe its own evolution. 

Quantum mechanics might have something to say about the idea, but we should not get attached to believing in it because we ‘want’ it to be true. Again, we should consider such ideas in a probabilistic way. Here I believe that in all probably it is not, but further research needs to be conducted. 

But whichever way it is,  if it does exist in some form, then it probably doesn’t have any opinions, and definitely does not think like a human, and have any concerns about what is good or bad, right or wrong, due to the fact that it only observes, observes the universe and itself evolving, where the sum of it will be everything including you, me, the mountains, the oceans, all of life, and the stars, every beam of light and every atom, so why would it want to complain about those things that inhabit the universe, those things that it is constructed from. It would be like you complaining about a specific neuron that takes part in a particular thought.

If we took a look inside your brain while you were being mindful of the momentary you, this, we would probably find that the neural circuits responsible for complex thought, was in a state of equilibrium.  Maybe, where all neural circuits, except where the observer resides (probably within the Thalamus/Insular/Cingulate cortical regions), then the firing patterns would probably be more homogeneous.

A mixture of skills

Humanity thrives upon the mixture of skills that society can bring to it.

Each individual possesses their own particular set of skills through the interactions of those neurons and neural circuits that thoughts are constructed from. 

Individuals within many other species are not so diverse, they’re far more homogeneous, where they only need to solve those problems particular to their species. 

If humans were like that, say if the human skill was only one particular variation, such as say carpentry, then we would not have been able to solve all of the other diverse problems that we do to construct the highly complex world that we have built for ourselves. A more homogeneous human race would not have been so intelligent. And in all probability, those neural circuits within this hypothetical human race would not have the diverse range of neurons and neural circuits that modern humans and humanity maintain.

Homogeneous synthetic intelligences

Just as human intelligence would be restricted by neurons and neural circuits being of a limited type and number, so are current synthetic intelligent systems.

Many current LLM’s are configured for language but find it difficult to solve problems of logic. There are many other areas where AI’s systems suffer, but these are soluble if neural circuits can be constructed in line with how the neurons and neural circuits interact in biological brains when solving such problems. 

Heterogeneous

The heterogeneous nature and intelligence becomes obvious when we consider the interplay of the environment and those beings that have to navigate that environment.

If we now turn to life itself, we can see that the more diverse attributes of the environment, each attribute of the environment that is of concern to the organism or system, needs to replicated within the neural networks of the organism or system.

Neurons, as all cellular systems, are largely constructed from proteins. This diverse range of chemicals are built from a range of twenty amino acid building blocks. There are an infinite number of ways in which these building blocks can be linked together, and once linked, folded into a diverse range of shapes. There are over 240 million currently known protein, while humans are estimated to have between 80,000 and 400,000 proteins that have a diverse range of functions. As can bee seen, this heterogeneous range of those building blocks required to build cellular systems, can create highly heterogenous systems, such as neurons and neural circuits.

 

From these understandings, we can see that for synthetic intelligences to perform as human like as possible, that they need to have collections of neural circuits specialising in the specific tasks that human brains and consciousness need to tackle, and the data that needs processing to achieve those goals.

Heterogeneous Biological Neural Networks

Neurons come in all sorts of shapes and sizes.

When we consider the homogeneous box on the last page, all boxes were the same, no differences.

If society was full of the people with the same skills, then not much would be able to be achieved, it would be as usefully intelligent as say a flock of say pigeons, all pretty good at being pigeons, doing pigeon things, pecking, flying, getting randy and courting, maybe finding their way home is about as intelligent as it gets, but not much else.

Humans are different, as are their brains. If we looked at each neuron, we would see that the shapes are all very different and diverse (except maybe grid cells in the entorhinal cortex, which allows us to navigate our environment, a bit like a blank paper, where all maps can be projected onto it). They need to be to encode the complex environment onto the neural networks that thoughts are constructed from. This heterogeneity needs to be so to encode the environment, but the processes within each neuron has to be more homogeneous, so that it functions as a system.

If you look around the body, you will see that other cellular organs are much more homogeneous, its only the brain that is so complex.  Some scientists consider that the brain is the most complex thing in the universe. Who are we to argue, many of us don’t even know the person that lives next door.

Heterogeneous Synthetic Networks

Should Synthetic Networks simulate the heterogeneous nature of biological neural networks

Currently, neural networks function through ‘simulating’ some of the properties of neurons, namely the interconnectivity. But biological neurons have many other functions, such as, oscillations, spiking trains, synchronisation and neuroplasticity . These aspects of biological neurons allow them to interact in many varied ways, allowing them to dynamically evolve, processing new types of data and to learn efficiently within a dynamically evolving environment.

Modern AI’s still function through calculating the strength of those connections through digital switching circuits. To calculate those other aspects of biological neurons would consume far more energy, but may allow for less emphasis on connectivity, while more emphasis on the dynamic and adaptable nature of those neurons, giving rise to greater intelligence while consuming less resources.

F.A.Q.

Lorem ipsum dolor sit amet.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.