This content was published by Andrew Tomazos and written by several hundred members of the former Internet Knowledge Base project.
The Human Brain and Computer as Cascading Computation
Welcome to the first edition of the Internet Knowledge Base newsletter.
We want to start by making a simplified comparison between a human brain and a computer.
++ THE HUMAN BRAIN ++
"The human brain is my second favorite organ." - Woody Allen.
Outside things act upon the human body. Light enters the eyes. Forces press upon the skin. Air vibrates against the ears. These outside forces cause long thin organic wires inside our body, our nerves, our sensory neurons, to send electrical pulses into our brains.
Our brains consist of about 150 thousand million neurons. If a neuron were the size of a toy marble, then your brain would be the size of a large suburban house. A few neurons accept external input signals, and are called sensory neurons. The ones that emit external signals are called motor neurons. The rest are interconnected in the middle.
Most neurons have connections with other neurons, about 10,000 on average. Inbound neuron connections are called dendrites and outbound ones are called axons. Where the axon of one neuron connects with the dendrite of another neuron is a small chemical soup called a synapse.
When the electrical signals arrive at the brain from the nerves they are converted to chemicals at the synapse. These then add or subtract the desire to fire on the neurons that follow.
When the total desire to fire on a neuron from the sum of its incoming connections reaches a critical level, it causes the neuron to convert the chemicals at the synapse into one big electric pulse and fires it down its axons. This pulse is then converted to chemicals at the other neurons' synapses, and the cycle continues.
What is interesting is that when a neuron fires, it (A) resets its synapses and (B) fires a short sharp pulse down all its axons. This is as opposed to firing a little bit, or a lot, or over a long period. Neurons do not fire continuous signals, varying the strength of the signal over time - instead, neurons send a pulse and then reset.
This storm of neurons firing continues all the time. This is what brain activity is. Eventually the neurons connected to our motor neurons fire, in turn causing some motor neurons to fire. Electricity travels down into our muscles. We walk, we talk, we type.
Each neuron in a way, takes thousands of results from other neurons, weighs them, then broadcasts the result to lots of other neurons. In a sense they both co-operate and compete.
150 thousand million of them are doing this at the same time.
Now lets talk about computers.
++ THE COMPUTER ++
An average computer, like the one from which you are reading this message, also has a brain in a sense.
Computers hold pieces of data in a variety of different areas. (Technical: RAM, ROM, hard drive, CPU registers, BIOS, etc)
People press keys on the keyboard, move around the mouse, generating inputs for the computer. These create electric signals that pulse into the computer which sends the signals inside it, making changes to some of these pieces of data.
From the time you switch your computer on, until you turn it off, a power source is used to send a regular electrical pulse throughout the system at many thousand million times per second. With each beat of this pulse, a big element inside your computer, called the processor, takes a piece of data, and uses it as an instruction. A small calculation happens or a manipulation on one or more other pieces of data. With the next pulse the processor takes another piece of data from the pool, and the cycle continues.
The manipulation the computer makes is based on data, and the data manipulated can end up being used to determine yet other manipulations.
This process continues until the data associated with output devices is changed. When it does, electric signals are sent from it. The picture on the monitor changes, the speakers emit a sound.
++ THE HUMAN-BRAIN & COMPUTER SYSTEM ++
The parallels are obvious. The outside world affects the system. Stimuli are converted to electric pulses, and sent into the system. The system makes thousands of millions of on-going cascading little manipulations using electrical pulses. Some of these end up leaving the system as output.
Now, when a person sits on a computer what is happening? The output of the computer is converted to light and sound at the monitor and speakers. This then travels into the eyes and ears of the person. Electrical signals enter the brain, are processed, generating signals which travel down to the fingers. We press keys and move mice. These actions in turn are converted to electrical signals, enter the computer, and so on. The human brain and the computer brain are part of one big system.
Everyone sitting at the computer using it, looks pretty much the same. They stare ahead at the pretty lights and twitch their fingers and forearms a lot. Imagine for a second if we take someone from 2000 years ago and show them someone using a computer. What would they think was happening here? Most likely they would think we were in some kind of weird trance. The fact is you can't see what is happening inside the system, any more than you can look inside someone that is thinking to himself. The human and computer are closed in a loop together.
At least they used to be, now we have invisible electrical pulses or electromagnetic radiation that connect the human-computer system to other human-computer systems, through networks and the Internet.
++ ADDITIONAL DETAILS ++
This is a simplified explanation of human brain function. There also are many incoming and outgoing neurons in the autonomic nervous system, a highly complex system that regulates the mostly unconscious, vegetative functions of the body, things like controlling heart rate and blood pressure, temperature regulation, digestion, sexual responses and the hormone secretion of various endocrine glands.
In fact, this cascading approach to processing is a very simplified example of human interaction or the underlying frameworks in which the mill of the universe itself grinds. Energy interacting with varying arrangements of itself. When you think about it, any interaction has a cascading effect on all future contaminated events.
Neurons use a traveling charge differential (a depolarization) in their outer membrane caused by a sudden ionic exchange between the inside and the outside of the cell. So what difference does this make? It means a neuron transmits information much more slowly than an electrical device. A neuron is capable of firing at a maximum speed of about once every 60th or 70th of a second, whereas silicon based devices such as microprocessors work a speeds millions of times faster. The massive parallelism of a brain is what makes it possible for it to compete with the microprocessor. But for how long?
So which one is faster, human brain or computer? IBM BlueGene is a very fast computer. It is currently being used to simulate a neural column at the molecular level. A neural column is a 1,000,000th chunk of the human brain. So if you look at it like that, human brains are 1,000,000 times faster than a computer at this moment. How long will take until computers catch up? About 60 years, in terms of raw processing power.
The brain also has some key self-modifying abilities which are both little-known and difficult to replicate in a computer. Neurons physically move around in the brain as you learn, making and breaking connections according to need. The Glial cells make up 90% of the brain and serve some part in our cognition. They were thought to be inactive and are the source of that old 60s quote "you only use 10% of your brain". More recently they have been demonstrated to take part in brain activity on a chemical basis rather than on a half-chemical half-electrical basis. Interesting, their effect becomes much stronger when any given stimulus is repeated 3 times.
Until very recently, what was thought to be Junk DNA (that part of the DNA which appeared not to do anything) actually becomes activated in nerve cells in the brain and central nervous system. That is, not only our understanding of genetics missing great swathes of information, but genetics appears to directly affect how our brains adapt. Since the genetic code is comparable to a program (or operating system?) written in trinary, there are parallels (sci-fi?) with AI systems written in self-modifying code.
Back to Index