The human brain has been compared to many things in an attempt to understand how it works. For instance, similarities have been pointed out between the connections in the brain and a large telephone switchboard. Since then, frequent parallels have been drawn between the brain and modern digital computers. As the twenty-first century begins, this comparison may be particularly appropriate because computers are being used in many applications that until recently were largely thought to be beyond the capability of machines and the exclusive province of the human brain.

The Computer Analogy

Computers are now being used in tasks from playing chess at a grand master level to visually recognizing what is being “seen” through video cameras in specific applications. The human brain’s comparison to a computer is the best analogy thus far, although it is important to note that there are important differences.

Digital computers are built from logic circuits or “logic gates” that produce a predictable output based on the inputs. The inputs are generally one of two selected voltages, which are generally thought of as the binary values 0 (zero) and 1 (one). Gates may produce a logical 1 output if all of the inputs are 1 (an AND function), if any of the inputs are 1 (an OR function), or if the inputs can be inverted, producing a 1 only if the input is 0 (a NOT

function).

From these simple functions more complex operations can be built by interconnecting logic gates, including circuits capable of adding numbers and memory devices capable of storing information. A digital computer may be built from millions of individual gates for its central processor and may use hundreds of millions or even thousands of millions of gates for its memory.

In the brain, there are special nerve cells known as neurons that function in ways that are strikingly similar to the logic gates in digital computers. Neurons possess structures for transferring signals that are lacking in most cells: a major fiber called the axonand a number of smaller fiber branches called dendrites. A neuron gets input from the dendrites, as well as from the central cell body, and subsequently produces an output signal in the axon. This signal then provides inputs to other neurons, or to other cells such as muscle cells.

Neurons are similar to computer logic gates because the output of an axon is a series of pulses of on and off signals, and never is a partial signal.

The inputs to the neuron determine if these pulses in the axon occur, and if so, at what speed they are repeated. The signal down an axon is an electrical signal, although the connection of signals between the axon and the dendrites of other neurons, known as the synapse, is achieved by a reaction of chemicals known as neurotransmitters.

Unlike logic gates, which are thoroughly understood, the exact operation of the neuron is still not completely clear. Science has probes that can measure the activity in a single neuron, but research is still ongoing to determine exactly how the neuron responds to potentially thousands of simultaneous inputs. It is known that while some signals to the dendrites help cause a neuron to  fire, others inhibit the firing process. But not all inputs contribute equally—some input signals to a neuron seem to have a stronger contribution than others. Repeated operation of a synapse may change its dendrite’s contribution to the output of the cell, and this appears to be a major part of the memory process.

Speed and Processing

Whereas a computer operates at very high speeds, making perhaps 100 million or more sequential operations in one second, the chemical transfer of information at the synapse is relatively slow, occurring in about one-thousandth of a second. A computer, however, must process information through one or a small number of central processors and execute its function or program in a sequence of steps. In contrast, the brain processes information in a parallel fashion, with large numbers of neurons analyzing signals and producing outputs at once.

Modern supercomputers are just beginning to conduct  parallel processing, with perhaps as many as a thousand central processors. It is not yet understood how to break many problems down into parallel operations, which the brain does automatically.

Digital computers operate by sequentially following a series of instructions in the computer memory, also known as a program, but the brain has no close equivalent to a written program. Its operation is based on linked networks of neurons that make up all of our mental processes, such as memory, cognition, emotions, intelligence, and personality.

Although it is frequently said that the average person uses only about 10 percent of their brain, it is hard to support this statement scientifically.

The brain, by its very nature, is redundant. A failure of a single logic gate may render a digital computer nonfunctional, but the loss of one, or tens,  or even of hundreds of brain cells may have little or no noticeable effect on the human brain. Whereas a small number of people show some remarkable traits, such as a photographic memory, it may be that most people record just as much in their memory, but their ability to access the infor-

mation is weaker, requiring more stimulus before the brain recovers the memory.

Whereas a great deal is understood about digital computers, far less is understood about the human brain. There is still much to be learned about how the neurons of the brain are arranged and interconnect, and how the inputs from our senses are processed. Rapid progress is being made in this area, but it is certain to be an important field of science full of new discoveries for years to come. 

______________________________________________________________________________________________

Reference

Crick, F. H. C. “Thinking about the Brain.” In Scientific American Offprint: The Brain,1984.

Hubel, David H. “The Brain.” In Scientific American Offprint: The Brain, 1984.

Tsien, Joe Z. “Building a Brainier Mouse.” Scientific American (April 2000):62–68.