Artificial Consciousness/Neural Correlates/Neural Models

Neural Models

When Hebb first proposed a Mathematical Model of the Neuron, he had a problem, it wasn't practical to study human neurons, so he was forced to study Giant Squid Neurons and hope that his experiments indicated something about human neurons as well. The nice thing about giant squid neurons, is that they were large enough to insert the electrical contacts of his day, and so he was able to at least get a sense of how neurons interrelate to each other.

After he had proposed the mathematical model of the Synapse he needed to define a formula for the neuron itself. He suggested a simple aggregate of synapses adjusted for the basic neural logrithmic function.

Marvin Minsky in his book Perceptron however, noted that this model was not accurate because there was a second order response factor not taken into account in the basic Hebb formula. Other Models were suggested, but each could be challenged relatively easily, if only because the models were too simplistic. To this day we still do not have a definitive neuron model, that captures mammalian neurons, let alone human neurons, but we are getting closer to understanding them especially now that we are beginning to learn about the cellular neuroscience at the micro-biological level.

Of recent interest, is the research being done on the connection between the synapse and the DNA which is involved in the growth of synapses and fibrils. While the biochemistry is still somewhat vague, the use of genetic KO techniques has opened up new vistas for understanding the complexity of long-term memory at the cellular level.

The following Models are a transitory look at our current understanding of neural function, undoubtedly future discoveries will change these models and skew them in directions we cannot predict at this time.

Neural Network Models of the Neuron