Abstract

This thesis formulates and evaluates a mathematical model from an engineer’s point of view based on the currently-known information-processing processes and structures of biological neurons. The specification and evaluation of the RealNeuron model form a baseline for current use in engineering solutions and future develop ments.

The RealNeuron is a carefully-reduced model that retains the essential features of more complex models. A systems engineering approach is used to formulate it, i.e. the model is described as using multiple resolution levels with configurable modular elements at each resolution level and is then implemented, verified and validated in a bottom-up method. It is computationally efficient and only adds or subtracts ion concentrations based on the states at the membrane structure’s level. The results are integrated at the lower levels of resolution. The RealNeuron’s simple calculations make simulations on personal computers possible by using standard spreadsheet software for a seven-neuron classical-conditioning neural circuit. All the simulated states at the highest level of resolution (i.e. pumps, channels, etc.), the intermediate levels of resolution (i.e. membrane potentials, neurotransmitters in the synapse, etc.) and the lowest level of resolution (i.e. conditioning signal, conditioned signal, conditioned reaction, etc.) are available on a spreadsheet.

The RealNeuron is verified in a bottom-up manner. The pumps, channels and receptors are verified first. These components are then integrated into the different membrane types (post-synaptic membrane, main membrane, axonal membrane) and verified while the membrane components are validated simultaneously. This process is repeated until individual neurons have been built up and RealNeuron networks have finally been constructed. The RealNeuron is verified and validated in configurations for AND, NAND, OR, NOR, NOT and XOR logic functions. It is also verified and validated by the implementation of classical conditioning.

In a noisy environment, the RealNeuron’s performance is dependent on the pump’s parameters in the main membrane of the sensor neurons.

This thesis proposes that a grade of machine intelligence is used to distinguish between the different synthesis requirements for intelligent machines.

An engineering synthesis of a RealNeuron network, based on classical conditioning, demonstrates how to implement a RealNeuron network that can be used in machines built to the grade of machine intelligence requirement which is classical-conditioning learning implemented with neural networks that can change learned associations in a dynamic environment.

Title Page

Towards an understanding of intelligence

For this discussion, intelligence is limited to the scope of biological intelligence as found in carbon-based life forms. This type of intelligence is found in nature of which human beings are very good specimens. This type of intelligence is what people normally have in mind when they talk about intelligence. A standard dictionary of the English language, [126 ], defines "intelligence" as:

1. Intelligence is the quality of being intelligent or clever.
2. Intelligence is the ability to think, reason, and understand instead of doing things automatically or by instinct.

The word "intelligent" is defined by [126] as:

• A person or animal that is intelligent has the ability to think, understand and learn things quickly and well.

In [126 ], "clever" is defined as:

• Someone who is clever is intelligent and able to understand things easily or plan things well.

From the above definitions, intelligence is the ability to think, understand, learn and plan things quickly and well instead of doing things automatically or by instinct. Based on Boole's classic work [11 ], the abilities to reason and think are combined.

[11] G. Boole, The Laws of Thought. Cambridge: MacMillen, 1854.
[126] J. Sinclair, Ed., Collins Cobuild Advanced Learner's Dictionary, English Dictionary. Glasgow: Harper Collins Publishers, 2003

The RealNeuron is a mathematical model from the viewpoint of an engineer that is based on the biological neurons’ currently-known information-processing processes and structures. It is a carefully reduced model that retains the essential features of more complex models. It is computationally efficient and only adds or subtracts ion concentrations based on the states at a membrane structure’s level. Its formulation makes it possible to optimise it.

Summary of RealNeuron

Results of eye-blink RealNeuron circuit

(c) 2006 Louwrence D Erasmus

Example of a RealNeuron circuit

(c) 2006 Louwrence D Erasmus

Here is an example of a RealNeuron circuit based on the description of the eye-blink circuit of a rabbit given by R. Menzel, Neuronale Plastizitaet, Lernen und Gedaechtnis. Neurowissenschaft Vom Molekuel zur Kognition. Berlin Heidelberg New York: Springer-Verlag, 1996, pp. 485 - 518

Introduction

This blog describes an exercise in complex systems modelling and simulation implementation. The information published in this blog is copyrighted by Louwrence D Erasmus and forms part of my Ph.D. studies at the Nortwest University, Potchefstroom Campus, South Africa.

Biological brains are one of the most complex systems known to exists in the universe. They are whole universes in themselves. The topic of brain modelling and building of machines that act as if they have brains, is a fascinating topic of study for centuries. The fact that biological brains try to model themselves through the centuries is another interesting philosophical study and outside the scope of this thesis.

The emphasis is on the modelling of the fundamental electro-chemical building blocks of intelligence as observed in nature. The ways how these building blocks interact in complex structures to achieve intelligent processing units, are also investigated.

In the end, a theory or model is useless if not applied. In engineering, theories or models are implemented to solve problems. The implementation of the model derived in this thesis is in a simulator that could be used in the design of products.