There is one biological neural network, which has not received the attention it deserves from mainstream science. This network is modular and is repeatedly utilized throughout the brain. The module has a pattern recognition framework, which facilitates the flow of information between networks, using a common language of internal representation. Each module has massive memories and powerful intelligence. While ignoring the mechanisms which facilitate the operations of this network, science has traced the paths along which these modules progressively integrate information in the brain.
In the primary areas of the cortex, these modules convert all sensory information into a common identifiable language of neural impulses. The recognition messages proceed to networks in secondary areas, which coordinate binocular vision and stereophonic sound. This integrated information travels to modules in the association regions, which recognize events. Recognition of events triggers complex emotionally supervised motor controls, which finally define all human activity. All these modules follow exactly the same intelligent processes, making them identifiable as the pivotal biological neural network of the brain.
The Modular Biological Neural
Network – A Revealing Definition
Neuroscience does not link the word “pattern recognition” to the word “network.” Wikipedia defines a biological neural network as a “series of interconnected neurons whose activation defines a recognizable linear pathway.” If the cells fire together, it is a biological neural network. But, if it is a “functional entity of interconnected neurons, which intelligently regulates its own activity using a feedback loop,” Wikipedia terms it a “neural circuit.” If it is intelligent, it is a circuit and not a network. The “circuit” label highlights the failure of neuroscience to identify the pattern recognition role of innumerable functioning biological neural networks in the nervous system.
The Modular Biological Neural Network – The Discovery
While its official definition suggests that a biological neural network lacks intelligence, a few scientists have already discovered the mechanism, which grants it a massive intelligence. A 2004 Nobel Prize was awarded for this very discovery. The olfactory system intelligently remembers, identifies and differentiates between subtle smells. The researchers had used calcium imaging to identify individual mouse receptor neurons, which fired on recognition of specific odors. They exposed the neurons to a range smells. They found that a single receptor could identify several odors. At the same time, each odor was identified by several receptors.
In the experiment, scientists reported that even slight changes in chemical structure activated different combinations of receptors. Different combinations of receptors fired to identify different odors. Neural firing combinations formed the internal representation of smells by the olfactory network. The scientists had discovered an intelligent biological neural network, which used combinatorial coding as its language of internal representation. Neural firing combinations were the language of internal representation for the olfactory neural network. The researchers believed that the taste network also followed the same internal representation system.
The Modular Biological Neural
Network – Combinatorial Power
The olfactory system contains over 10,000 receptors. Just 100 receptors could represent 100 x 99 x 98 x 97 x .... x 2 x 1 unique possible combinations. That represents more than 1, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000, 000 possible combinations! The recognition of combinations of firing at their dendrites can enable a single neuron to fire to identify trillions of unique combinations. Combinatorial coding has been discovered. It is a feasible language of internal representation. On this basis, similar biological neural networks can store vast memories and recall them in milliseconds. Such networks are the modules, which operate in all regions of the brain to empower the breadth and sweep of human and animal intelligence.
The Modular Biological Neural Network – The Visual Field
A hierarchy of neural networks finally recognize objects by analyzing the pixels of light arriving in the visual field. Individual visual receptor neurons fire in response to a small subset of stimuli within its receptive field. The neural firing combinations of their axonal outputs become the dendritic inputs of a neuron in V1. This neuron fires on recognition of a combination, which indicates a vertical line. Combinatorial firing from myriad neuronal network modules are identified at higher levels to indicate location, brightness, color, edges, line and curves. Each neuron can store a virtually infinite number of combinations of such aspects within milliseconds.
The human mind can recollect any one of 10,000 images displayed at 1 second intervals. The higher levels recognize and remember innumerable aspects of the millions of pixels of a single photograph. Neurons fire, when they recognize subtle combinations. At the highest levels, researchers discovered the “Bill Clinton neuron,” which fired on recognition of just one special face. The cell fired on recognizing three very different images of the former President; a line drawing of a laughing Clinton; a formal painting depicting him; and a photograph of him in a crowd. The cell remained mute when the patient viewed images of other politicians and celebrities.
The Modular Biological Neural Network – The Historic Process
Before the arrival of nerve cells, the earliest multicellular forms moved about and swallowed, or expelled food, by contracting their cells. The contraction was effected through chemical signals, the forerunners of hormones, which diffused quickly throughout the system. But the diffusion of chemicals was slow over longer distances and they could not be specifically targeted. Nature developed neurons to transmit specific information faster.
The entire process of neuronal interactions have been based on the pattern recognition model. Those networks in the early reptilian nosebrains recognized smells to be safe, or dangerous. The 2004 Nobel Prize describes the mechanism of those networks. The fine distinctions in the environment which they could make can hardly be explained through mathematical network models. Dogs can quickly sniff a few footprints of a person and determine accurately which way the person is walking. The animal's nose can detect the relative odor strength difference between footprints only a few feet apart, to determine the direction of a trail.
The Modular Biological Neural
Network – The Arithmetic Error
The reason why science still ignores the pattern recognition model is a single erroneous perception. That view prevents an understanding of the mechanisms of the biological neural network. The root perception of mainstream science is that neurons compute. Their standard explanation is that the axon hillock of a neuron triggers an action potential, when the arithmetic total of input signals received by its dendrites reach a specific threshold. Neurons are presumed to use some form of computation.
Science errs fundamentally with its assumption that mathematics initiates the action potential. Neurons do not compute. They recognize combinatorial patterns. The axon hillock of each neuron in the network recognizes specific patterns of the input signals at the synapses of its dendrites. On recognizing a pattern, an action potential flows down the axon of the neuron. This process explains the downstream acknowledgement by science that the omnipresent modules of the biological neural network described here performs clear pattern recognition functions.
The Modular Biological Neural
Network – Neuroimaging
There is overwhelming evidence to show that specific biological neural networks perform clearly defined functions in specific regions of the brain. The activation of particular brain areas, when people perform particular tasks can be identified through functional neuroimaging. fMRI (functional magnetic resonance imaging), PET (positron emission tomography) and CAT (computed axial tomography) have been extensively used to identify functional structures, or to assess brain injury through high resolution pictures.
Researchers have identified dysfunctional neurotransmitters such as dopamine in the basal ganglia of Parkinson's patients to yield insights into the networks, which cause specific cognitive deficits. Predictions of these deficits enable pharmacological manipulations, which deal with specific networks. There is clear evidence that the modular biological neural networks perform intelligent functions within the system. Yet, such intelligence is attributed to mathematical models, which fail to explain the powerful memories or the subtle intelligence of these networks.
The Modular Biological Neural Network – Hebbian Learning
The Hebbian learning theory suggests that the strengthening of active synaptic junctions could store network memory. Donald Hebb theorized that "Cells that fire together, wire together." He suggested that simultaneous activation of cells leads to pronounced increases in synaptic strength between those cells. He suggested that this led to “associative learning.” He suggested synaptic plasticity as a mechanism, which can store memories in networks. But visual memories imply changing combinations of neural firing at the same synapses. Each image is an arrangement of millions of visual pixels, arranged in marginally different combinations. A movie has wide screen images with millions of pixels changing 25 frames per second for 90 minutes! So, the process of watching a movie would strengthen ALL synapses in the visual system!
The Modular Biological Neural Network – Perceptrons
Instead of acknowledging the role of pattern recognition in biological neural networks, science offers several explanations of their neural mechanisms. All these explanations attribute various types of computation to form the basis for interactions between neurons. McCulloch showed theoretically that networks of artificial neurons could implement logical, arithmetic, and symbolic functions. Perceptrons, or artificial neurons are simplified models of biological neurons. While such models do carry out mathematical and logical computations, they cannot explain the phenomenal memory or the broad sweep of the human intellect.
The Modular Biological Neural Network – PKMzeta
Researchers suggest long term potentiation (LTP) as forming the basis for memories of neuronal networks. Dr. Sacktorat discovered a substance called PKMzeta, which was present and activated in neighboring cells with LTP links. The PKMzeta molecules formed into precise fingerlike connections among brain cells that were strengthened. The molecules remained in place to sustain the speed dial links, which enabled heightened responses to danger. However, when a drug, which interfered with PKMzeta was injected directly into the brain, the animals forgot their fear. The animals even forgot a strong disgust they had developed for a taste after the administration of the drug. It was hoped that by disabling LTP, the drug could blunt painful memories and addictive urges. The ability of LTP to handle urgent messages does not explain how the system remembers last night's dinner menu.
The Modular Biological Neural Network – Combinatorial Coding
All scientists are agreed that biological neuronal networks require a language of internal representation. The vast extent of animal and human knowledge requires the storage of memories, coded in this language. The language needs to translate external perceptions of the world into complex philosophical concepts. The system is known to have the ability to access any portion of this knowledge within milliseconds. Combinatorial coding can reasonably explain all these capabilities through the interactions of myriad biological neuronal network modules.
This page was last updated on 27-Jan-2016.
I really loved the self improvement plan post. Its great food for thought and the steps are actually actionable as compared to many other self help sites out there.
Joe Glen USA.
As a clinical therapist, I have found your site very useful!
I love it. ...
Andrew Montgomery USA.