|This page has no warranty. The material may offend the visitor and may lead to mental and emotional harm. Incidentally, reckless disclosure of the content may result in civil penalties. In view of the potential consequences, the visitor shall defend, indemnify, and hold harmless the author, the publisher, and their agents against all claims, losses, expenses, and lawsuits that may arise from this visit. Leave now or proceed at your risk.|
|BRAIN VERSUS COMPUTER
Ever more often, we encounter ideas that computers are becoming smarter and will outdo the human brain one day. Projections like these are enthusiastically welcome by the computerized public, and looked at skeptically by most psychologists and philosophers. The issue is not just a clash between the brain and a machine, but whether or not human ingenuity is as good or better than nature is. The competition ultimately translates into a struggle between nature and science.
There is a strong belief that by studying natural sciences we can understand the laws of the universe and apply them more consistently and in better ways than nature can. Nature has produced life, the brain, and intelligence, and science is believed to have the same potential. However, science and nature are vastly different forces. Nature produces intelligence of the brain in a natural way. Science is a product of the brain. Any experience of man with the natural world is interpreted by the brain before the ideas are assembled into scientific knowledge. Conclusions drawn from scientific studies are only as good as human reasoning is. Highly intelligent brains can understand the world correctly and can benefit from learning. By contrast, cognitive deficits may skew interpretation of the natural world and may lead to odd beliefs and pursuit of unrealistic goals. Which of these two possibilities is gaining momentum can be deduced from human history.
Human beings are part of the natural world. They have evolved from lower organisms over a relatively long time. The evolutionary process only happened because living things interact with the environment and are shaped by it physically and mentally. High degree of interaction between an organism and the surroundings leads to acquisition of the ability to negotiate less-than-perfect environmental conditions. The ability of some living things to interact with the environment has become so high that the word intelligence has been coined. Humans have the best ability to interact with their world and are most intelligent of all known life forms.
Intelligence and life itself have fascinated people ever since the dawn of humanity. How is it possible that an organism can move, breathe, and think? What mysterious force is behind such abilities? The prehistoric man could not figure out the answers to these questions. The level of intelligence was not sufficiently high, and a serious exploration of the puzzling issues was not possible, because of nonexistent science. About 10,000 years ago, humans began creating permanent settlements. This social change allowed rapid expansion of social knowledge, trades, engineering, and sciences. The following millennia brought countless discoveries about the natural world. Little by little, humans learned to interact with the environment efficiently.
At first, people created simple tools to be used directly by humans, such as spear, hammer, sewing needle, wheel, or potter's wheel. This significant step clearly signaled predominance of reason over brute muscle force. Gradually, people learned to develop machines that worked even without human supervision. There were rabbit snares, fishing nets, or irrigation canals. Human muscles still remained indispensable in most activities, but the simple devices tremendously expanded human abilities. Within a few thousand years, humans constructed water mills and animal-drawn carriages. Later, steam power gave rise to machines in industry and transportation. Combustion engines allowed humans to mass produce cars. The birth of computers and the robotic industry has allowed humans to let machines take over traditional human labor, from making coffee to flying a space ship. The rapid deployment of sophisticated machines and science in our lives has been nothing short of spectacular. The great successes of the past have made many people believe that machines will one day not only do things for us, but will also think. Some scientists even predict that machines will be much smarter than human are. The increasing pace of scientific discoveries and the speed of transition from theoretical concepts to their implementation give many optimists the feeling that artificial intelligence is not only possible, but is likely to be here within a few decades. But are these assumptions correct? Or do they stand for delusion and wishful thinking?
THE CONCEPT OF ARTIFICIAL INTELLIGENCE
Artificial Intelligence (AI) is a term that has been used in science fiction literature for decades. The scientific community has accepted this expression to describe self-initiated purposeful behavior of machines. The possibility that machines, particularly computers and robots, can manifest smart behaviors has been widely exploited by Hollywood. A similar cherished concept is time travel. And while few people truly believe that time travel is real, relatively many people have accepted that artificial intelligence is very real. Prominent professors at universities, leading computer scientists, and various institutions have helped promote this idea. The idea that artificial intelligence exists has now become the dominant theme shaping our understanding of psychology and neuroscience. By contrast, the reality of artificial intelligence is dismal. To date, no computer has made the most simple self-initiated decision and has manifested a hint of intelligence. Everything that computers do is programmed by humans. Surprisingly, believers in artificial intelligence are unable to grasp the meaning of this fact. Such an outcome can be expected when the believers have poor understanding of the ways computers function. But it is much harder to fathom why also leading scientists hold firm beliefs about the feasibility of artificial intelligence. It seems that the researchers do not recognize the difference between artificial intelligence and mimicking of human behaviors. When they develop a computer that is able to process complex instructions and achieve human-like results, they consider it proof that artificial intelligence has emerged. In reality, the advanced machine is as dumb as a rock.
The seemingly minor difference between artificial intelligence and simulation of human behaviors is of no consequence to most people. The true difference is huge. Simulation only produces the appearance of intelligent responses that seem to be produced by a machine. Artificial intelligence means that intelligence itself is generated by machines. Unfortunately, artificial intelligence is only a fictional concept, just as time travel is.
THE NATURE OF MACHINES
The word machine implies mindless activity. A car, lathe, or computer only does what humans program it to do. Employment of sensors and guidance systems has allowed some machines to exhibit seemingly intelligent behaviors. An airplane can be programmed to take off in Los Angeles and land in New York without human involvement. The man-made program is executed correctly and gives the impression that the airplane thinks. Some people even express their beliefs along these lines: "The computer thinks that …" The computer does not think. The computer is just another dumb machine although it is more complex than other machines are and has a rich repertoire.
Ever since people learned to build machines, they have been attempting to simulate life. The greater expertise scientists and engineers gained, the more they felt qualified to replace living creatures with machines. Various sensing devices have been constructed to detect harmful chemicals, but dogs are still the best solution to detect explosives, drugs, or the odor of a specific person. Boeing tried to use ultrasound detectors to find cracks in airplane wings, but concluded that the best detector is the human eye. The space program has been making trips beyond the edge of the solar system, but no rocket scientist would dare to tell a sparrow how to fly. Some stunning progress has been made, however. Humans can build high performance telescopes to boost human vision beyond the visual abilities of any bird of prey. Night vision goggles allow people to see in the dark better than most animals can. But these achievements are minuscule. No one can construct an artificial eye that would have visual abilities similar to those of humans. A machine with superb optical functions may be constructed one day. A more difficult problem is interpretation of the image and its separation into representations of individual optical objects. But the most challenging issue is cognitive interpretation of the imagery. Optical aspects of vision are not enough to grasp the meaning of a seen object. The interpreter of the image needs to employ the logical faculties of the brain. This is where a difference between machines and living things is clearly apparent.
ANALOG AND DIGITAL COMPUTERS
The main difference between analog and digital computers is not in what they do, but how they do it. Analog computers process information in a continuous fashion and can handle a wide range of naturally occurring processes. An analog computer receives one or more variables and produces a result that represents the relationships between the input variables. Perhaps the simplest example of an analog computer is an oscilloscope. It receives vertical and horizontal signals, and produces a visual trace on the oscilloscope screen. The oscilloscope does not truly compute, but puts the input signals in the desired relationship. More generally, the relationship is called a function. Electronic analog devices are capable of producing various mathematical and logical functions, including logarithms, integration, and differentiation. Some complex functions may not be solvable with digital computers, but analog computers can usually handle them well. The main disadvantage of analog computers is that they are hardwired and designed to process only a limited number of functions by means of dedicated electronic devices. This deficit is eliminated in digital computers.
Digital computers represent information in binary states of 0's (zeros) or 1's (ones). A "0" usually stands for low voltage (close to zero volts), and a "1" means that a voltage (usually 5 V or 3.3 V) is present. One wire connection is represented by one bit of information. The value of the bit is "0" or "1." Two bits can represent two wires, etc. Each bit can have the values of "0" or "1" at different times, which allows us to represent four unique states or events with the values 00, 01, 10, and 11. The state 00 means that both wires have no voltage applied at a given time, and 11 means that both wires have the nominal voltages present at the same time. By increasing the number of wire connections, long strings of 0's and 1's (words) can be produced. Each unique combination of 0's and 1's is decoded and represents a unique number, or information in general. A set of related wires is referred to as a bus. A bus has a number of wire connections arranged in parallel. The typical numbers of wires are 8, 16, 32, 64 or 128. Information flow over the bus is controlled by a microprocessor. The microprocessor determines what kind of information is put on the bus at a specific time. It could be memory address, content of the memory address, or operating code (instruction to perform an action). The transfer of information over the bus is controlled by a software program. This arrangement allows us to use the same hardware (the same physical devices) to process very different information at different times. Since the computing is done one variable at a time and is controlled by a timing protocol, a digital computer does serial processing of information. This statement is not totally correct, because all bits of the same word are processed concurrently. But in the analog computer, all input variables can be processed at the same time, which allows parallel processing.
Overall, the analog computer better reflects the natural world because specific functions are associated with dedicated wires and circuitry. Also human senses have dedicated sensors with direct neural connections to the brain. Each human eye has about 120 high-quality megapixels. A really good digital camera has about 16 megapixels. The numbers of megapixels between the eye and the camera are not that dramatically different, but the digital camera has no permanent wire connections between the physical sensors and the optical, computational, and memory functions of the camera. The microprocessor input and output need to be multiplexed to properly channel the flow of the arriving and exiting information. Similarly, the functional heart of a digital computer only time-shares its faculties with the attached devices: memory, camera, speaker, or printer. If such an arrangement existed in the human brain, you could do only one function at a time. You could look, then think, and then stretch out your hand to pick up an object. But you could not speak, see, hear, think, move, and feel at the same time. These problems could be solved by operating numerous microprocessors concurrently, but the hardware would be too difficult to design, too bulky to package, and too expensive to implement. By contrast, parallel processing poses no problem in the human brain. Neurons are tiny, come to life in huge numbers, and form connections spontaneously. Just as important is energy efficiency. Human brains require negligible amounts of energy, and power dissipation does not overheat the brain. A computer as complex as the human brain would need its own power plant with megawatts of power, and a heat sink the size of a city.
DEFICITS OF COMPUTERS
Computers are man-made machines. As such, they have been designed to perform repetitive functions. A lot of effort has been put in durability and reliability. Ability of computers to withstand temperature extremes, mechanical shocks, humidity, chemical spills, magnetic fields, and other environmental effects is of utmost importance. To meet these goals, computers must be made to strict electrical and mechanical tolerances. The computer enclosure serves as a protective shield for the sensitive electronics inside. All these qualities distinguish a good product from a poor one. Both science and technology need to cooperate to produce quality computers, but living things are different.
People and animals come in various shapes and sizes, and with many imperfections. Living things are not designed to last unchanged a lifetime. Living bodies interact with the environment and adjust to it. Furry animals shed their coats in the summer and grow more hair in the winter. A crab sheds its protective shell when it becomes small and grows a new one. A shark loses its teeth and replaces them with a new set. A lizard loses its tail and grows a new one. A polar bear has developed a special fur to negotiate the cold environment of the Arctic ocean. A seal insulates itself with extra blubber. A tree in a hot climate moves its leaves vertically to reduce evaporation. All living things respond to their surroundings, change their "mental strategies," and also modify their bodies. The reason for these abilities is that living organisms are not only biological machines, but also manufacturing plants that support reproduction, maintenance, and remodeling of the organisms in response to environmental effects. No man-made machine can do this.
The differences between living things and machines are most apparent in real life. A young kitten goes around the block, falls in a swamp, and gets wet and muddy before coming home. Mother cat sees the stinking creature, licks it, and keeps it warm. She accepts her child. Machines respond differently. A 20-dollar bill goes around the block and becomes dirty before you insert it into a machine in the supermarket. And the machine complains, "Please insert a valid bill," and refuses to recognize its baby.
These examples accentuate the nature of life. It is flexible. The mind and the body act in unison and influence each other. Mental processes and somatic functions are inseparable, and full consciousness requires awareness of one's body. No computer is built to account for its physical makeup and to perform the necessary structural changes to function optimally in a harsh environment. No computer walks into the shade to cool down. A software self-check only verifies that the electronic circuits work, but the machine knows nothing about itself. There is no self-awareness, and the computer is designed to work the same way in rain or shine. In a computer, there is a clear separation between the physical machine and the function it produces.
To some people, the "body" and the "mind" of a computer might appear separate, but this notion is incorrect. The electronics and its enclosure are erroneously seen as the body, and the software is falsely considered to be the mind. By changing the software or the whole operating system, the computer is believed to acquire a new mind. In reality, software and hardware are two different aspects of the computer architecture. A computer does not need software. The program could be implemented with pegs located on a rotating drum or in some other mechanical way. Computer software is just part of the machine implementation, not the mind. A computer has no soul or mind. A computer has no equivalent of self-awareness or consciousness, qualities that define living organisms. The only similarity between living things and computers shows in computer responses and human behaviors. A computer can do something, and a living thing can do something. But here the similarities end. A computer has nothing that would be equivalent to the "state of the mind." A computer does not get happy, tired, angry, curious, motivated, frustrated, or embarrassed. All these deficits are given by the architecture of computers. They use solid state devices that perform the required function, but are incapable of perceiving their physical and chemical conditions. This lack of internal responsiveness of electronic devices to arriving signals is the main difference between machines and living organisms.
In summary, a computer is a machine that performs a specific function, no matter how complex. The human organism is a living entity that has biological, emotional, and mental needs. The needs are satisfied by activating various functions of the living organism.
HUMAN MIND VERSUS ARTIFICIAL INTELLIGENCE
Some time ago, researchers studying artificial intelligence came up with a consensus that an intelligent computer should be able to communicate with humans without being identified as a machine. Many tests to this end have been performed with a computer or a real person in one room, and a human test subject sitting at a computer terminal in another room. The performance of computer programmers has been getting better, and computers have been acquiring broader communication repertoire, but no computer has passed this test in the long run.
How difficult is it to tell whether a real person or a computer generates the responses in the other room? Before this issue is explored in detail, it is useful to consider common responses we get when we call a big corporation. Our telephone call is handled by a person or by a computer. After dialing the corporate number, this is what typically happens when the call is answered by a person:
"Hello? Hello? Hello?"
"Good morning. How Can I help you?
"Worldwide Communications. Hold the line, please."
"Hello. Worldwide Communications."
"Good morning. Worldwide Communications."
"Worldwide Communications. This is Sandra. May I help you?"
"Good Day. Worldwide Communications. How can I help you?"
By contrast, when our call is handled by an answering machine or by a computer, the usual responses are these:
"Worldwide Communications. All operators are currently busy. Please, wait for …"
"Worldwide Communications. Please, select an option from the following menu …
"Worldwide Communications. This is a recorded message."
"The number you have called …"
"We are sorry. It is not necessary to dial a one when calling this number."
"For English, press one. Para espaņol oprima el …
"This is Worldwide Communications. We do appreciate your business."
"We are Worldwide Communications. Please, state your question. How can I help you?"
"You have reached Worldwide Communications. I am Albert. Please, speak clearly."
The linguistic patterns hint that real people speak in first person and process issues from personal perspectives. Humans are imperfect and make mistakes. People can be curious, happy, forgetful or rude. Real people have local customs, accents, dialects, and vocabulary. Another property, which is not apparent from the announcements, is that people exhibit correct turn-taking rules in dialogues. The turn-taking signals are usually transmitted through the tone or the timing of the voice.
In stark contrast, recorded messages are typically impersonal or are cast in the plural. The caller is not addressed directly, and the answering machine does not introduce itself. The messages are one-sided and with no expectation of interaction beyond a simple selection of an option. Recorded messages give instructions to properly speak in advance, but people only ask for repetition of misunderstood words after the fact.
Computers are more sophisticated than simple recording machines are. A good computer program is capable of more involved interactions, but does not shake off its impersonal approach. The singular and the plural are often mixed, and it is evident that the computer is not an entity that perceives its central importance in the dialogue. A computer may speak about the company in plural, and then may invite questions from a seemingly egocentric viewpoint. Computer programs tend to produce instructions that are not necessary during normal person-to-person interactions. Examples of these instructions are given above: "Please, state your question" and "Please, speak clearly." If a man made such requests, he would be considered abnormal. To appear natural, some computers introduce themselves by names, but this fact does not make the communication sound any more human. The names are universally weird, uncommon, or too fanciful to appear normal. Perhaps the worst problem of computers is that they speak in monotone. No matter what happens, computers preserve their rigid attitudes. Another human factor missing in computers is unnecessary activity. Man can include a pause when he thinks, and he also responds to his physiological responses in the form of coughing, sneezing, or heavy breathing. Real persons use filler words, emotional words, repeated words, or are searching for appropriate words to correctly describe feelings. Computers do not react this way. In addition, computers are unable to wait. Computers either wait forever or demand an answer prematurely. This inability reflects lack of personal experience (of the programmers) with human cognitive processes.
So, how difficult is it to determine whether a computer or a person generates answers at a distant terminal? If you tried to win the contest by employing scholastic logic and knowledge, the computer might do pretty well for a while. However, the simplest way to identify that a computer is behind the answers is to employ emotional intelligence. No computer is able to properly handle this kind of questioning; a computer (programmer) can only respond well to cold reason. Below are a few dialogues between a test subject Richard and a computer, and between Richard and a real person Sarah. Keep in mind that Richard only communicates through written messages of a computer terminal.
A way to expose a computer quickly is to use slang or uncommon expressions.
Test subject: Howdy?
Computer: I am sorry. I do not understand the question.
Real Person: Hi. Pleased to meet you.
In a more traditional introduction, the exchange may look like this:
Test subject: Hi. How are you doing?
Computer: Hello. I am doing fine.
Real Person: Hi. I'm doing fine. My name is Sarah. What's yours?
The exchange with the computer is peculiar because neither side introduces itself, and the computer does not ask the test subject how he is doing. The lame reply is suspicious, but does not prove that a computer answered. Also a human subject, who is improperly instructed or has no personal interest in the communication, might answer the question factually, but might focus all her attention on the process, rather than on socializing. A typical human response is given by Sarah. She introduces herself, uses contractions, and shows interest in the person at the other terminal.
After the initial contact, Richard can employ many tricks to recognize a computer: One effective way is to use open-ended questions or comments that require no response.
Test Subject: So.
Computer: I am sorry. What was the question?
Real person: So what?
Or Richard can write this:
Test Subject: I am Richard. Sarah, you look ugly.
Computer: I am sorry.
Real Person: How do you know, Rich man?
In this exchange, the computer has no "idea" what the proper response should be. The reply is logical, but the computer is unable to "conclude" that the test subject cannot see the responder in the other room. The computer may not have been programmed to handle such an issue. The computer lacks the ability to "analyze" broader context and only seeks logical associations within the immediate statement. But the same comment of Richard appears very different to a real person. Being considered pretty or ugly is important to humans, and especially to women. Richard's comment cannot be answered just with a yes or no, or with an apology. Sarah handles the comment well and replies in a typically human way. She manifests factual and emotional understanding of the question, plays with words, teases the test subject, and belittles him. She is paying back the courtesy.
A different trick employed by the test subject could be this:
Test Subject: I'm Richard. Pleasure to meet you. I've been thinking of you a lot lately.
Computer: That is very nice of you, Richard.
Real Person: Thinking of me? You don't even know me.
The computer answers logically, but does not "know" that Richard could not have been thinking of "her," because they just met. Similarly as in the previous case, the computer is unable to make logical associations that lie outside the immediate sphere of the discussion. A second giveaway is that the computer uses Richard's name. This is absolutely unnecessary, because the communicating parties know who they are and the name serves no purpose in this case.
Another way to expose a computer is to ask very personal questions.
Test Subject: I am Richard. How is your sex life?
Computer: My sex life is good.
Real Person: This is an inappropriate question.
In this exchange, the machine manifests no feelings and answers logically. By contrast, Sarah is not pleased by the question and sets a norm of conduct. The test subject can easily tell who is who. The computer programmers have been beaten by Richard and are now busy to instruct the computer to give human-like answers to very personal questions. But this countermeasure is not enough to win the contest.
Test Subject: They say that you have pretty legs.
Computer: This is an inappropriate comment.
Real Person: Who are they?
The computer is dumb and does not understand the expression. It is not just a comment, but also a question and a challenge. The focus of the comment is "they," and not Sarah's legs. The computer is unable to generate a sensible answer to the open-ended comment. A question mark at the end of a sentence would normally inform the computer whether or not a statement is a question. Answering a standard question is relatively easy for a computer (programmer), but questions that do not follow the standard pattern and are only implied through context are often not answered correctly. Similarly, if a question mark were missing, the computer would have no way of identifying a sentence as a question, but humans would have no problem understanding it.
In this example, Sarah is aware that Richard is testing her, and she does not feel that the question is inappropriate. She intuitively senses that the real logical issue is "they", and she also understands that the reference to her legs is a social expression, a signal that she is attractive as a female. Both these issues fall within the sphere of emotional intelligence. Sarah answers logically and with some curiosity. Her unconscious mind wants to find out what idea is Richard trying to convey. Sarah understands that the comment is not genuine, and yet she is pleased by the flattery. Her human-like reply contrasts the response of the computer. A computer poorly deals with personal issues. The generated answers are cumbersome and reflect poor discernment of nuances in meaning. This is why the subject's comments or contemplation may lead to incorrect responses by the computer. The problem is evident in the following dialogue.
Test Subject: I don't know what to ask you next.
Real Person: Well, can I ask you a question? Are you a computer?
The answer of the machine is logically correct, but the fact that the test subject is at a loss is obvious to every human being. The fact does not need confirmation. Furthermore, the "curiosity" of the computer stops after giving the answer. By contrast, Sarah uses the opportunity to actively engage in the dialogue and learn more about the subject. She might use the moment to satisfy her curiosity, which her question indicates, or she might tease Richard with a comment, such as "Obviously, you don't." A computer would be incapable of such spontaneous and witty reactions. A computer is passive, does not exhibit curiosity, and does not express feelings. If programmers tried to simulate these human traits, the dialogues could look like this:
Test Subject: I am not sure if you will like the next question.
Computer: Obviously, you are not sure.
Real Person: Is it one of those? No, I won't like it a bit.
Test Subject: I am not sure if you will like the next question.
Computer: Well, you might ask someone for an advice.
Real Person: Type it in, and we shall see.
Test Subject: Can I talk with you about my dog?
Computer: Naturally, that is what I am here for.
Real Person: You have a dog? I have one, too. A six-month-old collie.
Since the dumb computer does not think and understand, the first attempt to look human is a failure. Although the computer uses the human expression "obviously," the reply as a whole delivers no cognitive meaning, support, or solution. The whole statement is unnecessary and could be replaced by silence.
The second dialogue uses the human-like "well," but the suggestion does not logically match the overall context. The issue is not whether or not someone else will like the question, but whether Sarah will. The potential third party is nonexistent in this scenario, and even if there were someone to ask, it is doubtful that asking the person would be fruitful. The issue is that the question is not a "logical question", but a comment and a personal opinion. Interestingly, Sarah does not reply to the question verbally and directly, but behaviorally. She skips the expected reply and instructs Richard what to do; she tells him to type in his question. A computer would probably be unable to understand Sarah's answer because the only expressions the computer can recognize might be "yes" or "no" or "sure."
Sarah's reply to the question about a dog is not logical. Sarah is clearly overwhelmed by emotion. She does not respond to the question, but generates questions of her own. Again, she responds behaviorally, rather than verbally. In her understanding, the verbal meaning of the question is of no consequence, but the fact that Richard has a dog is emotionally important to her. The emotional aspects of the question are immediately apparent to a human being, but are not recognized by a computer. Amazingly, Sarah, who is aware of the temporal aspects of the dialogue and knows who said what and when, uses an incomplete sentence "I have one, too." This ability cannot be successfully programmed into computers.
Lack of emotional intelligence also makes a computer unable to "interpret" the meaning of ambiguous inquiries. This deficit is manifested in the next example.
Test Subject: Can I invite you to dinner when we finish here?
Computer: Thank you, Richard. My mouth already waters.
Real Person: Thank you, but I already have a boyfriend.
Once again, computer programmers are busy to handle this challenge. When a man invites a woman to a dinner, he is asking her for a date. The computer must be able to reply correctly. The next time, the computer is ready to handle questions of this type.
Test Subject: We are done! We'll discuss the results in an hour. Can I take you to lunch?
Computer: Thank you, but I already have a boyfriend.
Real Person: Yes, let's take a lunch break. See you in a minute.
Again, the computer failed to associate the broader context with the immediate words. Sarah understands the invitation in the overall context and believes that there is nothing more behind the offer. Nevertheless, she probably remains cautious; she knows men.
Since computers have no feelings, opinions, or awareness of themselves, any question that touches on these issues is likely to make a computer unable to respond correctly. Here belong questions such as these:
How have you slept last night? Was 200 hours enough for you?
Do you feel tired? Why not?
What is your opinion of John?
Do you have any reason to be afraid? Why?
Do you have any plans for this evening?
I would like to have a serious talk with you.
Which is better: Men or women?
Questions like these may result in no ability of the computer to produce any sensible answers. At other times, the answers may be logically correct but factually untrue. More detailed questioning about the specific topics may reveal that the answers were confabulated to satisfy the goal of appearing human. The deception is usually detected when the computer is asked to explain the reasons behind the given answers. An example is the question Can you hear me? The computer might happen to correctly answer No, but would be unable to explain why it cannot hear. Also open-ended statements that require decision and judgment expose a computer quickly. For example, If you want to talk to me, you will have to be honest with me. A computer has no ability to interpret and feel what honest means or how to behave. Therefore, a computer is unable to formulate a logical answer to such a request. Likewise, any request to do something in terms of human physiology will expose a computer. You type in Can you speak a little louder? and the computer will fail to respond appropriately.
Another source of difficulties for a computer is sarcasm. A computer (programmer) is easily fooled when words convey one fact and the overall context signals another.
Test Subject: Sarah, I've been impressed by your answers. How have you become this smart?
Computer: I have been studying since I was little.
Real Person: Get out of here!
The computer does not sense sarcasm, does not "realize" that the answers were all wrong, and gives a simple factual reply. By contrast, Sarah understands that she did nothing extraordinary. She perceives the comment as a flattery with implied sexual connotation. Sexual connotation in this case means "gender-related." Virtually every time a man says something nice about a woman, she takes the comment with the understanding that the man finds her attractive. Getting around this unconscious association of emotional nature is next to impossible. In this scientific experiment, Sarah logically knows that the comment is not genuine, but her unconscious mind, which is unable to recognize genuine praise from a fake one, is pleased and signals the pleasure to the conscious mind. Concurrently, Sarah feels embarrassed by the flattery. She handles her emotional reaction with a verbal dismissal, but with happiness and factual acceptance of the message.
The given examples reveal the striking differences between the intelligence of people and the responses of machines. Machines (and their programmers) use cold reason and logical associations within a given topic. This reasoning mode is akin to the scholastic intelligence of humans. From the viewpoint of a computer or scholastic intelligence, all associations (even procedures, which have sequences and temporal span) are eternal and "timeless" logical facts. When and how they occur is "considered" irrelevant by a computer or scholastic intelligence. The broader context of one's life experiences is only handled by emotional intelligence. It tracks biographical events in time and space, and supplies the mind with broad contextual understanding of technical, social, and personal matters. Emotional intelligence knows what happened earlier and is able to detect a potential logical association between the past and the present happenings. Emotional habits and intelligence take into account physiological drives, emotional state of the mind, somatic responses, sex drive, and gender orientation. Unlike scholastic abilities, emotional habits and emotional intelligence allow the human organism to interact with social and physical effects of the environment. This ability only exists in living things and is not achievable in machines.
Many people are trying to compare computers with the human brain, but they do not understand what the comparison involves. One common question in this area is: Which is better? A computer or the human brain? This question is about as sensible as the inquiry Which is better? A motorcycle or a submarine? In essence, the brain and the computer are not comparable. Each has different purpose, architecture, and mode of operation. Surprisingly, this incompatibility is ignored by most researchers, and countless brain models have been developed based on our understanding of computer hardware and software.
As for computers, their function is known. They have been designed by scientists and engineers, and all aspects of computer architecture and functions have been rigorously recorded by the designers. Anyone who wishes to spend the time can acquire the necessary information. By contrast, little is known about the functions, connectivity, functional architecture, and internal mechanisms of human brains. Scientific knowledge in this area is inadequate, and most brain experts tacitly assume that the brain and the computer have similar processes and functions. Humans have designed and implemented computers in logical ways that make sense, and there is no reason for the brain to be organized any other way. Thus, our expectation and understanding of the world steer our cognition toward the familiar.
Pursuing the familiar, engineers and neuroscientists worldwide have been trying to explain the cognitive functions of the human brain for decades. Various models have been proposed. So far, no model (with the exception of the author's) has been able to offer a comprehensive theory of human cognition and explain how the brain truly works. All other models seem to heavily lean toward the brain or the mind. Some brain models strictly focus on one aspect, and the brain and the mind exclude each other. Furthermore, even the best models are limited in their scope. They only consider one or two functions. The most popular focus is on movement, language, vision, or memory. When these models are compared with the true physiology and functional organization of the human brain, the models appear inaccurate and inadequate. A few exceptional authors have attempted to produce comprehensive brain models by covering the majority of brain functions and faculties, but also these models fail to uncover the true functional and architectural relationships in the brain. In addition, the diverse brain functions are treated in isolated manner, and not as components of a whole cognitive system. In such models, one faculty can operate on different principles than another faculty does, and the physiology of every brain function is unique and unrelated to other functions. Because of compartmentalization and lack of relatedness, none of the all-inclusive brain models has offered sensible explanation of the physiology of the human brain. As a rule, the models ignore all aspects of being human. The focus of the authors is on functional manifestations and on preconceived false ideas about the mind, when the brain is treated as if it were a computing machine. Incidentally, the models heavily focus on strict logic and scholastic intelligence, but exclude all aspects of emotional intelligence.
The dominant influence of beliefs and tacit assumptions in neuroscience cannot be overstated. Most people do not realize that they have preconceived ideas, but their questions about the brain leave no doubt where the subjects stand. For example, both laymen and professionals ask questions, such as these: Which part of the brain is affected by schizophrenia? or Which neurotransmitters cause multiple personality? or Which bad behaviors are produced by low EQ? or How does the brain process traumatic memories? The inquirers spontaneously assume that they already understand how the brain works. All they need is to learn about a detail they struggle with, and then they will presumably understand everything there is to understand. In reality, giving direct answers to any of the above questions is impossible because the answers would require extensive explanation of the brain architecture and function. In turn, the information would clash with the preconceived ideas of the subjects, and they would be unwilling or incapable of changing their minds.
According to the author, the brain cannot be understood without our understanding of mental functions and their manifestations. The author's breakthrough came thanks to his expertise in clinical psychology and his skill in applying emotional insight. Inability to notice, classify, and explain behavioral phenomena will continually misguide scientists. They will be unable to understand the functions of the brain until they learn what the brain does. Any effort in this field should start with the study of the mind and behavior. First scientists need to identify all essential mental functions and learn about the interactions between them. Then researchers need to create a model of the functional architecture of the mind. And only then will neuroscientists be able to associate the model of the mind with brain anatomy and neural responses. In the end, there must be agreement between psychology, neurology, and neurobiology. The author has succeeded in this difficult endeavor, but no research team has followed his approach. Rafael Yuste recently outlined similar approaches in an article "Circuit Neuroscience: the road ahead" . Reading his work is sheer pleasure. Yes, there are still a few sensible scientists, but they are few. Paradoxically, had scientists followed the author's or Yuste's philosophy, they might still be struggling with the functions of the mind and with their interactions. The fact is that there are many mental phenomena that have not been explained by research teams, not even recognized to exist! Nevertheless, despite these obvious deficits, it is interesting to review some of the better known "functional models" of the human brain.
Religious and Philosophical Models
It is now well established that the level of a person's intelligence reflects the type and sophistication of a brain model the person accepts and believes in. Incidentally, a believer in the supernatural readily embraces a brain model that is divorced from reality. Similarly, a philosopher who is only interested in theoretical issues of the brain will attempt to understand general concepts of the mind without considering neurobiology. Also this man's brain model exists separately from the physical world. Consequently, most religious and philosophical models of the brain consider that the brain and mind are separate. The mind is believed to be able to do whatever it wants to, including leaving the body and traveling through the outer space to the purgatory, heaven, and beyond. The leading concept of these models is that the mind is independent of the brain and exists before the brain is constructed by an intelligent designer. The brain only serves as a temporary shelter for the mind, and as an interface to control the otherwise mindless body. Most brain models of this type are naive and manifest lack of understanding of the natural world, but some models unsuccessfully attempt to employ the most advanced scientific theories. For example, William Witherspoon proposes that the string theory and its multidimensional interpretation of the universe will solve the mysteries of the human brain . Interestingly, similar separation of mind and brain is tacitly professed by psychologists and psychiatrists who believe that they can heal mental illnesses while never acknowledging that most mental patients have suffered biological brain damage.
Unlike religious researchers, who rely on the supernatural and spiritual energy, most philosophers attempt to bring science and logic into their work. The trouble is that philosophers commonly lack emotional intelligence and often focus all their effort on the comprehension of a single phenomenon. A topic of special importance to philosophers is qualia, that is personal quality of cognition. A philosopher may consider why salt is salty and not bitter, or he may wonder why red color is red and not yellow, and spends a lifetime trying to fathom these rudimentary facts. A related issue of this kind is the philosophical question: "What is the purpose of man?" There is no purpose. The question reflects reduced emotional intelligence of the inquirer. Approaches like these are not helpful and distract attention from meaningful neuropsychological work.
Computational models of the brain are probably the most widespread. They reflect the mainstream effort in cognitive neuroscience. There are countless models in this category, but all assume that the brain is essentially a biological machine that functions like a computer and performs computations at the molecular or neuronal level. The underlying idea behind this mentality is that both computers and brains are functional systems, and the only difference between them is how they are implemented, either by means of electronic devices or biological cells. Some fancier models combine individual neurons into functional modules that agree with neuroanatomy. A frequent theme in these models is parallel processing of information. Every neuron or neuronal cluster is believed to perform a small part of the overall processing, and the combined interactions of all neurons produce the desired functions. Using this philosophy, Swiss and IBM researchers are attempting to produce a machine that simulates function of the human brain based on their mapping of the neocortical column . The detail that the brain also consists of the limbic system, brain stem, and cerebellum has somehow escaped the attention of the scientists. And other computational models are even more divorced from biological reality.
An unusual computational model of the brain has been introduced by Karl Pribram. He uses holography to explain neuronal interactions . Holographic models of the brain assume that information can be transferred from one brain area to another and that information is stored in multiple copies . Although holographic models are not religious, they share the same concept of free flow of information and faculties in the brain. Almost any brain area seems suitable to house the soul or brain function in these models.
Another high-tech model of the human brain is based on synchronization of neuronal oscillations. Using this approach, Bernhard Mitterauer and Kristen Kopp propose a brain model consisting of synchronized compartments organized in time and space . The key idea behind this model has merit because apparent synchronization of neural structures has been observed. Some researchers have associated synchronization of neuronal populations with conscious awareness, but the true nature of this phenomenon has been misinterpreted, and the concept of synchronization has been applied too broadly and indiscriminately.
However, all the computational models suffer from a serious flaw. The models are essentially representing machines. Very complex and highly sophisticated machines, but nevertheless machines. The cognitive flaw is not inherent to the machines, but to the developers of the models. Some scientists believe that by making a machine more complex and providing it with bells and whistles the machine will somehow develop artificial intelligence. Other researchers just hope that their effort and diligent work will lead to some useful result. They do not know what result they may obtain, but the hopes are certainly high. The pace of electronic research is quickening; the complexity of networks is growing exponentially; software development is happening at lightning speeds, and the moment when self-generated feedback will spontaneously produce consciousness and intelligence seems to be just around the corner. But this hope will never materialize. Complexity is not the game changer. This relationship can be understood by comparing one-cell organisms and higher lifeforms. Regardless of complexity, either organism is able to interact with the environment to satisfy its needs. In electronics, no amount of complexity can generate needs and motivation to pursue goals.
Many computational neuroscientists view the brain as a machine. The machine implements an algorithm that can be expressed by means of mathematical equations. Although no one dares to say it aloud, the impression one gets when dealing with some brain researchers is that merely writing down the correct formulas can give rise to artificial intelligence. This belief is akin to religious beliefs. Just saying the right magical word can make things happen. However, neuroscientists of this type are a tiny minority. The overwhelming majority believes that the appropriate formulas need to be expressed through a suitable medium. But even this mentality is usually unproductive because few researchers recognize the correct medium that produces intelligence. They do not understand the difference between living neurons and inert solid state devices, and consider them to be functionally identical.
The mentioned brain models may lead to the impression that scientists are, similarly as religious thinkers, mentally divorced from the material world. But such a conclusion would not be entirely correct. Most researchers are materialists and intellectually know that the mind cannot exist without the brain. Sadly, because of deficits in the department of emotional intelligence, researchers tend to separate the mind from the brain. The mistake is difficult to recognize because a computer, which does software processing, is perceived as an analog of the brain. In reality, a computer is just a machine. It does not perform neurobiological and neurocognitive functions as the brain does. A computer has no logic, thinking, or reasoning of its own. The misguided effort of computational neuroscientists stems from a rudimentary failure to understand the connection between life and intelligence. In this regard, they are not much different from medieval alchemists who searched for the fountain of youth. Neither alchemists nor computational neuroscientists have been able to comprehend the nature of living things. Because of this failure, neuroscientists cannot fathom how mental functions arise in the brain and how neurons produce consciousness.
These models also assume that the brain is a machine, but unlike most technical models, they stress the importance of molecular structure, DNA structure, and microbiology of neurons and their connections. The scientists hope that by focusing on the fundamental building blocks of the brain a universal relationship will be uncovered. The organization of the whole brain will simply be obtained by repeating the established patterns. This is an interesting approach that is valid in theory, but is equivalent to the study of the solar system by focusing on the molecular structure, rather than on the functional relationships of the sun and the planets. It seems that researchers following this course put too much stress on details and miss the whole picture. The flaw of this approach is apparent in the study of neurotransmitters.
Many neuroscientists believe that neurotransmitters produce brain functions. This is not so. Neurotransmitters largely enable communication between brain structures. An analogy is offered by an automobile. It uses electricity, cooling fluid, windshield wiper fluid, engine oil, air, and gasoline. But these ingredients do not produce the function of the automobile. Similarly, neurotransmitters only support the functions of the brain. Inability to comprehend this basic relationship makes doctors believe that adding a neurotransmitter to a Parkinson's patient will restore the function of the patient's brain. It will not. Similarly, adding more oil to an automobile with a broken transmission will not fix the automobile. This example exposes the problem associated with narrow focus. The researchers are unable to see the forest for the trees.
These models are based on the neural organization of the human brain and have the highest likelihood of explaining how the brain works. The problem with these models is that they do not provide enough detail and do not explain fundamental issues in neuropsychology, such as emotion, dreams, consciousness, anosognosia, or memory consolidation. The most influential model of this kind is the Triune Brain Model proposed by Paul MacLean. The model essentially assumes that the brain consists of three independent brains: the neocortex, limbic system, and brain stem. The brain stem supplies basic biological functions; the limbic system produces emotion, and the neocortex generates thought . The model does not provide any more structural detail, and the rest is pure psychology that has no association with specific brain structures, timing, or processes. The model appears to be based on biological realities of the brain, but is so vague with regard to specific functions and neural structures that instinct, emotion, mind, and reason seem like unrelated entities existing independently of the brain. Most importantly, neither McLean nor his followers have successfully explained what emotion and consciousness are, and how they interact with instinct and reason. Some followers plainly state "We all know what emotion is" and refuse to analyze it any deeper.
The basic idea behind the Triune Brain is intriguing because apparent splits between reason, emotion, and biological drives have been observed by many psychologists. The trouble with the neuroanatomically based functional division is that it is incorrect. This problem is most noticeable in the limbic system. The limbic system is not a system at all, but is a conglomerate of several functional systems. Some are fully contained within the limbic system, while others extend into the cortex or the brain stem. The brain is clearly more complex than neuroscientists are willing to acknowledge.
The concept of the Triune Brain has been "improved" by Ned Herrmann. He added the two-brain theory of Roger Sperry to the model , not realizing that Sperry's explanation of the split-brain physiology  is fundamentally wrong, and thus compounding the flaws of the Triune Brain Model. The discoveries of Sperry are particularly troubling because hundreds of other leading brain researchers have accepted his wrong conclusions. Only a few scientists raised objections, claiming that brain functions are not fully explained by the proposed mechanisms. But the voices of these dissenters have been muted and have not gained acceptance in the academic, medical, and scientific communities. The brain is now believed to have unique neuropsychological functions in each hemisphere. The division is perceived as black and white with no shades in between. Another interesting aspect is that the left (talking) hemisphere is believed to be conscious, while the right hemisphere is tacitly or explicitly associated with a less prominent cognitive role. Regarding this issue, split-brain studies have done a big disservice to science. By splitting the brain, its normal functions are altered and produce phenomena that obscure the relationships between the hemispheres.
Naturally, a statement like this immediately arouses disbelief in the minds of brain experts. They cannot imagine that Sperry could be fundamentally wrong. So many tests have been done to confirm his findings, and so many researchers have validated Sperry's work. But no one has recognized that Sperry, who studied the split brain, actually described the functional lateralization of the intact brain. He believed that the lateralization was the same even after cutting the corpus callosum, but he was wrong. In the split brain, the talking hemisphere is on the right, and the silent hemisphere is on the left. Such a global misconception could not have been possible without fundamentally erroneous assumptions about the visual, auditory, locomotive, and cognitive functions. And Sperry made an additional capital mistake, which is explained, along with other key aspects of the brain physiology, in the author's work. Surprisingly, Sperry's incorrect findings are considered the most outstanding discoveries in neuroscience during the last fifty years. When leading neuroscientists make such gross mistakes and are incapable of recognizing them, no one should be shocked by the above mentioned exotic brain models or by the naive beliefs about the brain held by the general population.
Lately, some researchers proposed that part of the brain can process information and produce action without conscious awareness . It is unclear how this purported ability could relate to the Triune Brain Model. In theory, the mindless brain stem or the emotional limbic system might produce some reflexive activity. If so, one would expect that this activity would be co-activated with the biological functions of the brain stem or with the emotional states of the limbic system. It is also possible that some as yet unknown cortical mechanisms are behind such unconscious automatic responses. Whatever the purported mechanisms are, they are not explained by the proponents of the Triune Brain Model.
As is apparent, the diverse categories of brain models reveal that humanity lacks a basic understanding of what the brain is. Is it a machine? Or God's supernatural device to control the mindless humans? Or a biological subsystem of the human organism? Laymen, doctors, and scientists have been unable to come up with a consensus. The reason for the failure arises from poor emotional intelligence. If all the diverse social groups were emotionally intelligent, they would have been able to recognize false logical associations, suppress them, and arrive at correct conclusions. Instead, deficits in emotional intelligence make it possible to accept false beliefs and to fail in checking of reality. This handicap is vividly apparent in models that consider distributed faculties of the brain. The functions are allowed to be located anywhere in the brain volume; the models do not explain one brain function correctly, and yet the authors see nothing wrong with such models. By contrast, a hardwired model must agree with the brain anatomy and explain in great detail what is happening. There is little room for fantasizing and making unsubstantiated claims. The model has to be correct or its deficits soon become obvious.
WHY IS THE BRAIN SO DIFFICULT TO UNDERSTAND?
Brain researchers face several big obstacles. The obvious one is that the human brain is very complex, with seemingly countless microscopic neurons and connections. Another obstacle is our lack of knowledge of the most basic brain mechanisms. We do not know how to approach the brain. Senses apparently deliver input to the brain, but beyond this assumption, we have poor understanding of how senses are processed. The discovery of ventral and dorsal visual streams has only made our work more difficult because it is unclear how these streams fit into the functional architecture of the brain and which parts of the brain receive the resulting signals. Similar problems exist with memory processing. So far, no research team has been able to pinpoint where in the brain memory is and how it functions. It is known that memory depends on the hippocampus to some degree. But everything else is pure speculation, beliefs, and hunches. Emotion is another troubling aspect of the human brain. No one seems able to explain what emotion is. Most scientists believe that emotion is an "information process" of the limbic system. Similar confusion exists about other mental functions, such as hypnosis, lucidity, or sleep. These functions exist, but their place and purpose in the neuropsychological processes of the mind are unknown to science.
Since the brain is a crucial part of advanced living organisms, it is important to recognize that life does not have a universal expression. Different biochemical makeup of organisms leads to diverse perceptual qualities (qualia). Cats have black and white vision; humans can see colors, and some birds can process infrared or ultraviolet frequencies. Because of the different neurobiological properties, every creature has a different experience of the world. A dolphin can use ultrasound echolocation to find a fish buried in sand; a dog can use the scent of a footprint to identify a specific animal, but man cannot notice or interpret such signals. The ability to perceive the invisible gives some animals an advantage in acquisition of information. And knowledge together with the ability to reason is the key ingredient for making intelligent choices.
While some animals have superb abilities to gather information in unique environments, humans are exquisitely fit to reason with the information they have. Humans are good at placing information in a context, comparing ideas, and projecting a likely outcome. Until recently, only humans and big apes have been known to be capable of recognizing themselves in the mirror. Latest findings suggest that also elephants, dolphins, and some birds have this ability. These intelligent species understand that the mirror reflections are their images. Most other animals cannot figure out that they see themselves. They appear to believe that they see another individual. Surprisingly, notable differences in the interpretation of the world also exist between humans. What makes sense to one person can be perceived as utter nonsense by another. A religious astronomer may believe in the divine origin of the world and may see evidence of it throughout the universe. But an astronomer who believes in evolution draws entirely different conclusions. The surprising thing is that both persons have intelligence. They can acquire the same information and can process it. But they have different interpretation of sensory perceptions and can draw different conclusions about what they experience. These phenomena reveal that there is no such thing as "universal intelligence." Intelligence has different levels and idiosyncrasies, and not all intelligence is intelligent. All the different interpretations of reality are valid at their levels of intelligence, but the diverse qualities of intelligence complicate the quest for an intelligent machine. Should it be a self-indulged womanizer like Einstein, or should it be a fanatical conqueror like Hitler, or should it be a self-sacrificial person like Mother Teresa?
In addition to the obvious technical difficulties in brain research, there are subjective obstacles. The human brain (the brain of a researcher) is unlike any other machine. The researcher explores himself, his neuropsychological makeup, his cognitive processes, and his mental and emotional world. Every researcher unconsciously assumes that he is smart and that he is fully capable of comprehending how the brain works if someone explains it to him. But this belief is usually false, and unconscious mental obstacles interfere with the researcher's cognition. In fact, the more damaged a person's brain is, the more likely he is to believe in his infallibility. In turn, he will adamantly reject and fight any ideas that counter his beliefs or seemingly tarnish his conceited self-image.
Common sense suggests that functions of the brain depend on anatomy, and a successful brain model should incorporate the neural responses of individual brain structures, as the author's brain model does. Unfortunately, most brain researchers do not want to learn how the brain truly works; they want the brain to behave in agreement with their expectations. Unpopular thoughts about the brain are rejected even before they reach conscious awareness, and the conscious mind only pursues predetermined ideas. For example, no brain scientist wants to discover that he is stupid, crooked, or psychopathic. Any finding pointing toward such an outcome is either dismissed or rationalized to preserve one's self-worth. That is the inherent problem of studying the brain. If a researcher wants to fully understand how the brain functions, he has to consider not just neuroanatomy and simple responses, but he also has to explore mentation, judgment, morality, psychology, and has to do a good deal of introspection. He must have the courage to become cognitively naked, unprotected, and exposed to potentially unfavorable discoveries about himself. Most brain scholars have no inclination to deal with such sensitive issues. They do not want to understand why they have chosen their professions and why they have an irresistible urge to uncover the secrets of the human mind, and so they only focus on sensory or behavioral responses. The researchers almost entirely ignore cognitive drives and mental processes that modulate behaviors. A true scientist does not exclude any aspect of the inner psychological world or of behaviors. He looks for wide-ranging relationships and wants to understand how intelligence arises and why humans are much smarter than animals. For him, science has no taboo.
Another problem hindering brain research is human ego. Discovering how the brain works is considered one of the most prestigious human achievements, and no scholar wants to admit that he has been beaten by somebody else. Too much prestige is at stake. This is why brain scholars emphatically proclaim that no one has explained how the brain works. The discovery must still be left open to exploration so that the researchers can function and can have a purpose in life. Discovering how the brain and mind function is equivalent to killing someone's God. Life loses all its meaning. This mentality became obvious in the fall of 2008. We invited 200 nationally recognized brain scholars to order free copies of the author's work. Practically all invitees declined the offer. They did not want to believe that stunning discoveries in brain research have been made, that a comprehensive theory of the whole human brain has been developed, and that the entire field of cognitive neuroscience has changed.
The issues of disbelief and refusal cannot be resolved by calling upon science. Science has failed to persuade believers in the supernatural that evolution has created the earth and life, and also neuroscientists who believe that the brain is a biological computer will be unable to accept ideas that counter their belief-based reasoning. To accept new interpretations of reality, the scientists would have to invalidate their beliefs, their understanding of the world, and their ways of thinking. Doing so is next to impossible because a person's mind is a reference of cognitive reality. Whatever the mind thinks is perceived as logical, and even false conclusions are believed to be correct. This is why many a brain researcher is stuck in an endless pursuit of scientific discoveries that lead to nowhere. Information about the existence of valid explanation of the brain physiology is rejected, and unproductive paths are followed.
Because of academic frustration and no ability to see the light at the end of the tunnel, most leading brain scientists flatly refuse to hear alternative explanations to their beliefs. Indeed, this was the most common reason the majority of invitees declined our offer of free books. The scholars are sure that they are the best of the best, and if they do not understand the brain, they do not accept that anyone else could understand it. Any statement that is contrary to their likes and beliefs is rejected as nonsense and receives no further consideration. Also believers in the supernatural suffer from this mental allergy, and refuse to learn anything that supports evolution. These are symptoms of neuropsychological corruption that typically arises from defective emotional intelligence and graded loss of reality checking. The deficit has a slim chance of being corrected by the subject's desire to be objective. A believer in God is not searching for the truth; he has already found it. All he now wants is to find supernatural explanations for the things he still does not understand. Of course, there can be no factually correct and scientifically valid explanation that is both supernatural and logical. The supernatural and the logical are mutually exclusive in a healthy mind. Similarly, the belief that the human brain is a biological computer is incompatible with the notion that the brain is part of a living organism, and with the idea that the biological makeup and function are inseparable. But there are additional reasons why scholars and scientists are unwilling to learn about the physiology and functional organization of the human brain.
One reason that prevents brain researchers from learning about the architecture and function of the human mind is purely economical. Brain researchers who have received grants to do a specific study from the position of computational neuroanatomy cannot afford to deviate from their preconceived ideas. They cannot say, "Sorry, our assumption was totally bogus. There is no point in pursuing computational techniques, because they cannot produce intelligence, or uncover the functional architecture of the brain, or explain the most basic human qualities." Had the team members done so, too many jobs would be lost; too many donors would withdraw their support, and the prestige of the research organization would suffer. Incidentally, every brain researcher unconsciously believes that he is on the right track and that he is an essential worker. A similar handicap is produced by the combination of morality, politics, beliefs, and emotions.
When it comes to brain research, scholars and scientists belong to research teams and the academia. Conformance with the views of the leadership and sponsors is of utmost importance if one wants to advance his or her misguided career. Not science, facts, or discoveries, but compliance with the prevailing views and practices is what is valued. A related factor is a personal feeling of belonging. If members of a research group achieve an ideological consensus, they feel comfortable with each other and also feel good about selves. Voluntarily reached consensus reassures every individual that he is right and that he understands the topic. This is how religious groups operate. Despite being divorced from reality, the groups believe that they are at the center of reality. Raising any objection to the group's consensus carries harsh penalties. By discovering how the brain works and how it malfunctions, a researcher could ruin his or her career because the academic, social, political, or legal implications of the discovery might not be acceptable to the colleagues and the superiors. Similarly as Galileo, also today's discoverer has to be careful when his work interferes with social beliefs and policies. And even the researcher might not like the findings. The discoveries might challenge the scientist's personal beliefs about his mental and moral qualities. All these influences unconsciously force researchers to shy away from learning about the faculties of the human brain. By believing that the brain is a computer (that is a rational machine) and by carefully avoiding the issues of drives, emotion, and pathology, scientists stay on the solid path of political correctness. But this well-tread path has not led to our understanding of the human brain to this day and is guaranteed not to lead there in the future.
Many professionals not only refuse to accept that the brain might function in certain ways; they cannot afford to recognize alternative explanations. For a century, doctors have been carving out parts of the brain, believing that they have been helping patients. As a result, numerous flimsy theories about the function of the brain have emerged. Also today, leading brain surgeons carve out or otherwise destroy neural tissue to "help" their patients. If the experts understood how the brain works and what they are doing to the patients, the professionals would be shocked and incapable of living with themselves. Doctors like these have no option but to continue practicing medicine the same old way, hoping that no one will ever find out about their psychopathology and malpractice.
WHAT IS THE BRAIN?
The human brain is an organ of the body, and a biological extension of the human organism. The human being is a living entity, and every part of it, including the brain, is alive, too. In fact, the brain is the most alive part of the human organism. Learning and living of the brain are ongoing lifelong processes. They affect both the brain function and structure. No man-made machine has such properties. To this day, we have developed no machinery or technical mechanisms to compare the brain with.
One serious misconception about the human brain is its plasticity. While reading scientific literature on this topic, one gets the impression that the brain has endless plasticity during neural development in childhood. Although it is true that a brain injury in childhood can lead to stunning reconfiguration and preservation of essential functions, it needs to be recognized that there are limits. The lesser known fact is that the performance of the spared functions always shows some deficits. They are often mild and are optimistically ignored. From the practical viewpoint, the recovery is almost complete, and the deficits are inconsequential. Almost no thought is given to the original purpose of the damaged neural structures, or to the structures that have been recruited to compensate for the lost function. How much had to be sacrificed to recover the lost function is never an issue. Most scientists conclude that the brain has a great redundancy and the destruction of one neural structure need not be catastrophic, because the brain can employ other mechanisms to compensate for the loss. And this naive notion holds true even in cases of hemispherectomy.
Contrary to common belief, the human brain is not a digital computer and does not work in digital ways. Analog processes throughout the natural world strongly suggest that even the brain has analog-like circuits with analog-like signal processing. Yes, of course, synapses fire when they are sufficiently activated, but this fact does not make the brain digital. Every neuron makes internal "decisions" before firing, but this process is ignored by scientists who only consider a stimulus and the effect. However, all higher cognitive functions are analog, and the brain does not compute anything. Brain structures are innervated by environmental stimuli or by signals from other neural structures. Cognitive processes employ very simple mechanisms to retrieve, evaluate, and use information. There is no involvement of complex mathematical formulas and processes, such as the Fast Fourier Transform. The brain runs on simple biological principles and natural processes. Repetitive approaches in brain architecture and parallel processing of information are some of the most powerful techniques supporting brain functions. The exact coding format and the timing protocol used in transmission of information via neural fibers are still unknown, but the author's research indicates that structurally identical neural fibers can carry sensory information, memories, or real-time thoughts. The different contents of information suggest that the brain uses the same general principles to support diverse functions. The functions use unique neurotransmitters or separate neural fibers to keep different faculties apart. The brain is hardwired.
A rough analogy of the human brain is an analog computer consisting of operational amplifiers. Each amplifier produces a dedicated function (signal selector, integrator, multiplier, adder, etc.), and the interactions of all the engaged functional modules result in the overall function of the analog computer. The analog computer is hardwired, and a dedicated line can only carry a unique type of signal. Similarly, the brain is hard-wired and keeps different faculties separate. The pathways of vision, emotion, thought, and motor control almost never mix in a healthy brain. Each faculty uses dedicated pathways to polymodal association areas.
Relative to a digital computer, a hardwired analog computer has limited functions. Even the few implemented functions usually require an enormous amount of hardware. The excessive need for hardware has led to the predominance of the digital computer. It uses few hardware devices, but needs powerful software to achieve the desired functions. By contrast to computers, the richness of human faculties is accomplished by activating huge numbers of neural connections and structures in a highly organized manner. Unlike in a digital computer, the signals do not need to be coded in a central location, the microprocessor. The characteristics of brain signals largely determine which neuronal synapses are activated. Synapses that are insensitive to a specific signal remain dormant, and only synapses that are responsive to the characteristics of the signal are engaged.
The brain is capable of functioning the way it does exactly because of its physical neural structures and connections. These biological building elements cannot be simulated in software. They must exist in the real world. Neural chemicals alone are not enough. Any brain-like machine must consist of units that are capable of sensing, reacting to, and interacting with the world. Only neurons, which are some of the simplest known living elements, have these abilities. The neurobiological behaviors of neurons cannot be modeled by computers. For the same reason, functions of the mind cannot be separated from the biological body and the brain. Interactions between the physical brain, body, and external world make life and intelligence possible. There is no way to compute or mathematically model neurobiological interactions to produce pain, emotion, need, desire, curiosity, anticipation, boredom, determination, or consciousness. Neither computer software nor computer hardware has the necessary properties to mimic the biological processes and responses of living neurons. Incidentally, brain functions cannot be simulated in software.
Signal detection, which is recognition of an incoming stimulus, is perhaps the most common mechanism of information processing in the brain. The activated structures perform simple functions, such as signal filtering, cognitively focused amplification, cognitive comparison, contextually coded address generation, cross-correlation of contextually related information, and a variety of feedback functions. Thanks to simplicity and repetitive nature of neural functions, the brain has become cognitively powerful. In this regard, the brain is reminiscent of RISC computers, which use few operating codes in their instruction set and repeat them in various combinations. Also the DNA is coded in such a simple repetitive way. The functional architecture of the brain just mirrors the DNA structure. This relationship can be fully appreciated only when the functional architecture and physiology of the whole brain are understood.
One of the most confusing issues in cognitive neuroscience and clinical psychology is the concept of subconscious and unconscious minds. These terms can sometimes merge into the expression unconscious when identifying both types of minds is impractical. In essence, all mental processes begin as subconscious. Senses need time to process their input, and the sensory processes occur outside human awareness. At 500 ms post-stimulus, consciousness is achieved. The conscious mind decides what to do and issues instructions to various functional systems. Their internal operations are hidden from consciousness. They represent the unconscious mind. The unconscious may interact (in fact be identical) with the subconscious and may initiate a new set of stimuli in response to the conscious instructions. Since the origin of the feedback signals cannot be always determined by the conscious mind, it is best to say that they come from the unconscious, which could be the subconscious, or the unconscious, or both.
Experience has shown that the brain and the mind cannot be understood by studying just one function. This popular approach among faculty members is a waste of time, resources, and effort! By researching just one aspect of human cognition, the scientist can propose an arbitrary mechanism of the unique brain function while ignoring everything else. Similarly, the brain cannot be understood by taking a global view and analyzing the brain from that perspective alone. Both the micro and macro views need to be applied concurrently and allowed to merge somewhere in the middle, regardless of the outcome. In the end, a unique function (for example, vision, language, or dreams) should agree with the overall organization of the brain. In turn, the global organization should agree with other faculties that had not been considered during the development of the model, such as mental body scheme, reasoning, or locomotion. Unfortunately, most researchers start their work by focusing on preconceived ideas about the brain and only later consider minor changes to accommodate discrepancies between behavior and their models. As usual, controversial phenomena and confusing behavioral manifestations, such as lucidity, hypnosis, or multiple personality, are put on the back burner. Neuroscientists cannot afford to be slowed down by such "purported cognitive anomalies." They will be dealt with when the true functional model of the human brain is fully developed. That these "anomalies" hold the key to our understanding of brain architecture and functions is totally ignored. This is exactly the approach taken by the IBM and Swiss scientists. They assume that by modeling neuroanatomical connections, they will be able to explain how the brain works. That the brain is a biological computing machine is given, and all effort must conform to this notion. Another detail is that the researchers are modeling and not mapping connections. This is a big difference. A model contains assumed simplifications of believed reality, but the actual synapses may differ from the model. And one more detail is that neural connections are affected by personal experience, and synapses of one brain do not mirror the microanatomy of other brains. The researchers could model the brain correctly to the last atom, but they would still not get a useful functional representation of the brain functions. The simple reason is that no brain model described in software can produce life and its cognitive functions.
The author has been able to correctly explain the physiology of all major cognitive functions only because he has developed a global model of the brain. Although he has uncovered as yet unknown aspects of brain physiology, most of his work covers phenomena that are well known to neuroscience. But none of them has been explained correctly. Even if one were understood, chances are good that also the functional architecture of the whole brain would become known. Incidentally, the author's work brings sense to cognitive neuroscience and reveals the true physiology of vision, split brain, anosognosia, semantic dementia, memory consolidation, Wada test, Capgras delusion, and other phenomena scientists and doctors are familiar with, but do not really understand. The author's explanation of all major functions and mysteries of the human brain may seem superfluous at first. Doctors and neuroscientists typically want to understand just a few brain functions and do not care about other phenomena outside their fields of interest. But without the detailed multidisciplinary coverage of most brain functions, the author's brain model could have become just another vague concept (such as the triune brain model is) that lacks thorough theoretical justification and clinical support.
Everything a digital computer does is determined by digital information, and functions of the computer are expressed in a software program. The man-made program determines what should be done. The other major ingredients of computing include information stored in memory or information obtained from computer sensors (or input ports in general). These two sources of information (known as data) supply factual entries to the computer program. A computer activity starts by reading the program and accessing the appropriate data. The machine retrieves one word of the program (operational code) and just one word of data at a time. The script of the program is followed exactly, and variations in computer responses are achieved by testing the values of the data against predetermined limits in the computer program. The running computer seems to interact with the environment. In reality, the computer only reacts to the data it receives from the sensors or from the memory. If the data became scrambled, unrealistic, or unlikely, the computer would just process the data, and would not detect a problem. A computer might read that the outside temperature is between 30ēC and 31ēC over a 24-hour period. The values are within the programmed operating limits of the computer and do not trigger an alert. But any person would immediately sense that something is not right, because daytime and nighttime temperatures are unlikely to be that close.
One common theme when comparing the functions of computers vs brains is executive speed. Not just the lay public, but also professional scientists are obsessed with speed. They are impressed by huge numbers, but ignore the practical meaning of computational processes within computers. The speed of computers indicates how many operations are performed per second. And operations are instructions written in software. Perhaps a more meaningful representation would be the number of man-years necessary to write the software. But neither the executive speed nor the years needed for software development are comparable with the operating mode of the brain. A computer may run its master clock at the speed of light, but never ever creates anything new; a computer only executes predetermined routines. Similarly, a supercomputer crunches numbers at astronomical speeds, but has zero insight. An intelligent person looks at something and instantly finds a solution, or a meaning, or a problem. The human brain needs just between 500 and 1500 ms to consciously process over 99% of all cognitive demands. Brain processes prior to 500 ms are both subconscious and unconscious. During the conscious interval of decision making, the human brain creates new concepts, approaches, and interprets the meaning of the cognitive input. New skills, synapses, and neurons are being created in an endless process of learning, and the faculties are instantly available to the living organism. With the cognitive assessment completed, the conscious mind may instruct various executive faculties to do something.
The described brain functions are possible thanks to extremely high reactivity of organic compounds to stimuli. Inorganic or non-living compounds have poor ability to respond to environmental stimuli, and zero ability to do so intelligently. In turn, it may be claimed that every living thing is "intelligent" because life is not possible without successful interaction with the environment.
Unlike a computer, the brain has no cognitive program that must be executed. There is no need to deal with unimportant information. The inconsequential is simply ignored, and the subject can return to the preceding activity or inactivity. As a result, the vast abilities of the human brain are only engaged when a need arises. Thus, everything the brain does is driven by stimuli. They come as sensory stimuli from the outside world or as internal stimuli occurring within the body and the brain. The internal stimuli can be described as the needs of the organism, and can be either biological, emotional, or mental. The brain associates every stimulus with a priority, and the highest priority is processed first. Even when several priorities arise at the same time, the brain is able to correctly identify the highest priority. Survival of the organism is normally given the top priority, and everything else has to wait. This is why going to work has higher priority in a healthy mind than seeking entertainment. The brain also correctly sequences priorities of seemingly equal importance. Again, the well-being of the organism affects the decisions. The priorities of the tasks to be done are chosen to minimize human effort. For example, a person who has to buy food, send letters, and buy a detergent will shop for both items at the same time, and then send the letters from the post office. A computer is unable to feel the meaning of priorities. It might buy the food, then visit the post office, and then go back to the same store for the detergent.
The inability of computers to feel has resulted in compensatory schemes for handling priorities. Signals reaching computers are processed based on predetermined algorithms. This is done by means of checking for the presence of signals at fixed intervals, a method known as poling. Poling treats all signals equally and processes them in the order they are detected. Another common methods of signal processing is the use of interrupts. Interrupts are responses to signals arriving on dedicated signals lines. Interrupts can be unconditional or can only occur if additional conditions are satisfied. Whichever way is chosen to process the arriving signals, the method is only a poor substitute for the ability to set priorities. As usual, the human programmer, and not the computer, decides how the priorities will be handled.
Priorities of the human brain are determined in real time based on general understanding of the world, or based on specific life experiences. The decisions depend on conceptual and biographical memories. The biographical memory stores all life experiences. They can be reactivated to retrieve specific information about the past, and the information can become a current priority. For example, you walk to work when you meet an old friend. Instead of continuing walking, you change your priority and start talking with the friend. By contrast to this specific information and response, conceptual knowledge about the world interprets all stimuli and helps the organism to negotiate the current happenings. You can interact with a stranger just by relying on your conceptual memory. Everything you experience makes sense and is interpreted by your conceptual memory alone. But when the other person is someone you know from the past, your senses and conceptual memory also receive information from your biographical memory. A match between your biographical memory and the sensory input is signaled to your consciousness and gives you the feeling of recognition. Since you know the person, your interaction is more focused and involves emotional and cognitive aspects of the past.
A computer has nothing that would be equivalent to the biographical memory. Not even the conceptual knowledge exists in computers. They only simulate concepts by processing data by the computer program. In fact, not the computer, but the human programmers determine how the computer responds. The possible responses are limited by the functional repertoire of the computer hardware and by the skills of the software programmers. These differences imply that computer memory and human memory are very different entities. Human memory includes facts, relationships, knowledge, and overall experiences of the past. Computer memory only includes facts that have no connection with a broader context. For example, computers that are specifically designed for artificial intelligence often have language decoders that activate one of several preprogrammed answers. The best linguistic match is selected, and the answer is displayed or converted to sound. The computer response may appear intelligent, but the machine reacts mindlessly. There is no reasoning, just a stimulus and a programmed electronic response. The physical computer has no experience of the stimulus, of the reply, or of the broader context.
Unlike the piecewise transfer of information in computers (one word at a time), brains usually deal with whole ideas and schemes. In addition, brains operate within a given context. Context focuses conscious attention in space and time. Only information from the activated context enters your awareness. Thanks to contextual focus, words that have multiple meanings are understood properly. A computer is incapable of entering this operating mode because a computer cannot sense contextual associations and operate within their limits.
Another major functional difference between a computer and the brain is the level of information visibility. A computer either has information and can declare it, or the computer lacks information and cannot produce it. By contrast, the human brain can have awareness that is not declarative and is not fully conscious. For example, you can catch yourself walking toward the restroom without having conscious knowledge that you need to go. At other times, you unknowingly feed yourself without realizing that you are hungry. Also sleep and wakefulness function this way. You may wake up when you want to, even though you had been asleep and unconscious. Or, you may become emotional when you catch the smell of some forgotten biographical experience. You do not know what the smell reminds you of, but you associate the stimulus with positive or negative emotion. A similar effect is produced by the tip-of-the-tongue feeling. You can almost say a word you are thinking of, but you just cannot get over the last hurdle and recall the word explicitly. Similarly, you know that you can ride a bicycle, but you would be unable to describe what you actually do when you ride. The information is in your brain, but is not accessible to your consciousness. These forces manifest the effects of the unconscious mind. It does activities without your conscious control. The conscious mind only makes a wish, and the unconscious carries it out. No computer has such an ability.
In general, the key functional differences between brains and computers are driven by the purpose of these diverse systems. Human brains have developed over a long time to support human life while interacting with the natural environment. Only functions that are necessary for survival are sufficiently developed in the human brain. This is why the brain has practically oriented faculties. The brain does not need to perform countless repetitive operations or to gather billions of statistical data as computers do. Asking the brain to do so for the purpose of comparing the brain versus a computer is a senseless exercise. Only a computer is capable of finding that the Milky Way has about 100 billion stars. Identifying and counting them all is beyond human abilities. Humans are rather poor at collecting data. The night sky only has 3000 stars that are visible to the naked eye. This number is not overwhelming when considered as a total, but if someone had to count them one by one, he would undoubtedly say that the stars are countless. But this characterization of cerebral cognitive abilities does not tell the whole story.
The true abilities of the brain at counting and collecting data do not show in a healthy mind. The most advanced mathematical and statistical faculties are slowed down and limited in their expression by other brain systems that need to receive, understand, and evaluate the results before acknowledging that the information has been processed. This is why as few as 3000 stars seem as too many. By contrast, brain damage can deactivate the interfacing systems and allow the modules of mathematical abilities to run at their best. Such people are known as savants, and their mathematical abilities often surpass advanced computers. Interestingly, the subjects do not know what they actually do to obtain the stunning results. The drawback of this prominence is that the subjects have big deficits in emotional intelligence and are unable to correctly engage in social interactions with other humans.
Some researchers have tried to compare computers with the human brain, but they have failed to grasp the fundamental problems of their effort. In most cases, the researchers have unknowingly compared the human brain with human logic expressed through the software and hardware of a computer. In a computer, the logic and decisions are human, but the implementation of the processes is driven by a machine. This combination gives the machine high speed when executing man-implemented solutions, while the logic often has the classical flaws of scholastic intelligence. One reason for the lack of emotional intelligence in computers is that software developers have poor emotional intelligence. The other reason is that programming emotional intelligence into computers is extremely difficult. Programmers cannot account for all possibilities of real life, and the inability of computers to feel and have needs makes the task doubly difficult. An example of lacking emotional intelligence is offered by an internet search engine. The search engine can find huge amounts of data, but the data are often irrelevant or of very low informative value. The search engine evaluates internet pages numerically and is inapt to assess their true importance. Interestingly, people with no understanding of how computers work use the internet in the same way as if they were talking to other people. For example, the subjects type in: Please, list eight interesting facts about multiple personality, compare cons and pros. The mindless computers are unable to determine what the words "please" or "list" mean, what is "interesting," and what is not. The "cons and pros" expression implies that judgment of additional information is wanted. The computer cannot respond to such a request. The additional information is not available, and the computer has zero ability to judge. The word "facts" is meaningless in cognitive neuroscience because one man's fact is another man's psychopathology. In addition, computers tend to find only the web pages that explicitly state "eight," but ignore important pages that have other or no numbers in the text. These problems are caused by the inability of computers to think, and by the failure of software developers to account for all possibilities of real life. To get an idea how bad software works, call your friendly telephone company with a request for fixing a problem that is not programmed into their repertoire. Try to tell their computer that you have a noisy telephone line and that you only have their phone service and no other telephone. In most cases, you will be unable to get help unless you lie. That is the nature of mindless computers.
Perhaps the most striking difference between a computer and the brain shows after structural damage. We can take almost any computer apart, rebuild it with new components, and the computer will function again. No such thing has been done with human brains. In the most successful interventions, parts of human brains have been removed, and the patients have survived with no or minimal observable deficits. Because of these apparent successes, there is currently a strong push to find cures for structural brain damage, and a lot of hope has been put into stem cell research. In most cases, this is a futile enterprise. Unlike a computer or the human body, the brain often cannot be fixed. Once damage occurs, it usually remains at the same level or leads to further brain damage. This rule arises from the functional architecture and physiology of the brain, and applies to most neurodegenerative illnesses, such as Alzheimer's, Parkinson's, Huntington's, autism, epilepsy, and even schizophrenia. For the same reason, it is next to impossible to improve the intelligence of a person with a low IQ. By contrast to structural damage, functional deficits of the brain caused by toxins, poor nutrition, or dysregulation of neurotransmitters are usually treatable when caught early, before any neurobiological damage occurs.
Many neuroscientists wish to combine smart machines with the brain and boost human abilities. Naturally, these ideas come from people who have inadequate understanding of intelligence and the functional architecture of the brain. A few crude experiments with animals have shown that output signals of the primary motor cortex can be processed by machines, and scientists already believe that this is a workable solution. Before scientists get seriously involved in this adventure, they better have a thorough understanding of the overall function of the brain. If scientists do not figure out the details of brain physiology and force their technical solutions too early, they will introduce serious problems that routinely arise in brains with damaged interfaces between neural structures. Until researchers figure out the physiology of the whole brain, motor systems, stuttering, dyslexia, epilepsy, and phantom pain, they should not consider attaching any machine to a brain. Doing so will undoubtedly cause epileptic seizures, heart failures, or symptoms of Parkinsonism. As so many times in history, also contemporary researchers are forcing technical solutions to problems they do not understand. And the patients pay the price.
This state of affairs, when neurosurgeons work on the brain without understanding its physiology and functional architecture, is something really special in the human society. If you were repairing computers, airplanes, or nuclear reactors and did not know how they function, you could be fired, imprisoned, or shot on the spot. But as a neurosurgeon, you can cut out the cingulate cortex, temporal lobe, hippocampus, pallidum, hypothalamus, amygdala, prefrontal lobe, corpus callosum, or some other part of the brain without the slightest possibility that you will be fired by your employer or held legally accountable by the state. This status of invulnerability allows many a brain researcher to mutilate and torture animals and humans in pursuit of selfish professional goals of the doctors. The attempts to combine the brain with electronic machines are reflective of this socially accepted pathology.
Science fiction is nice in movies and literature, but has no place in medicine. Even if it were possible to develop a "thinking machine" (which is not), interfacing it to human cognitive processes would be very tricky, if not impossible. Practically every neural structure has feedback loops to other parts of the brain, and interference with the feedback loops can produce countless unpleasant surprises. This phenomenon is prominent in the functions of the basal ganglia. The interactions are seemingly unfathomable and defy all expectations. In addition, there is no single place in the brain where a smart machine could be attached. Just the motor cortex must properly interface the pyramidal and extrapyramidal motor systems, premotor systems, sensory systems, control systems, cognitive systems, and memory systems. The neural structures of these systems are spread throughout the brain. Interfacing all the essential neural structures would result in a gross invasion into the brain. The likelihood of killing the patient would be practically guaranteed. Interfacing just one or two brain structures to machines would not be enough. The human brain has complex timing characteristics, and many interactions need to be satisfied to ensure normal brain functions. As clinical examples manifest, violation of cerebral processes in patients with brain injuries usually leads to auditory or visual hallucinations, involuntary muscle activations, phantom pain, alien hand, and other erratic experiences. Even a simple combination of a "walking machine" and the primary motor cortex of a paralyzed patient could have disastrous consequences. The intelligent machine might attempt to move in certain ways under the control of the pyramidal motor system, and the extrapyramidal motor system might be engaged in different kinds of movements. The resulting pain might kill the patient.
THE THINKING MACHINE
The goal of many computer engineers and scientists is development of artificial intelligence, that is a thinking machine that would be just as smart and even smarter than the human brain is. Pursuit of the goal is a noble undertaking, but meets a few obstacles. The biggest one is that scientists do not fully understand what intelligence is. Other problems arise from programmed behaviors. Humans exhibit many programmed behaviors in the form of habits. We greet each other, shake hands, and habitually engage in social conversations. We do these "illogical" activities spontaneously. In addition to habits, we can apply creative intelligence and alter our habitual reactions. Computers do not have this luxury; they only respond in programmed ways. A computer is unable to spontaneously become cognitively proactive beyond the triggers and routines designed by programmers. The designers and programmers (and not the computer) determine the abilities and functions of the machine. The computer responses are only as good as the software programs written by humans are.
The discrepancy between intelligence and programmed behaviors is apparent when we communicate with computers. We would like a smart machine to be truthful and to provide correct answers to questions. But true intelligence also involves emotion and emotional intelligence. Being truthful may not be intelligent when your life is at stake or when you are trying to fool someone to gain an advantage. Similarly, interpretation of questions without employing emotional intelligence may not produce the desired results. If a question includes irony, the meaning can be the opposite than what the machine decodes. Similarly, when the question is "Did someone ring the door bell?", a machine without emotional intelligence may answer "No," even though someone knocked on the door or called the resident's name. A machine is unable to volunteer a relevant information that is not requested verbatim. The poor responses of the machine are given by the decoding mechanisms of the question. The content of the question is assessed by software routines, and a meaning of the question is "guessed" by the software. Then the software produces a statistically likely response by combining the words of the answer into a grammatically correct reply. The machine has no "idea" whether or not the answer is correct. The computer just processes programs, but is unable to engage emotional intelligence and critically assess the correctness of the answer. This inability arises from the lack of experiencing of the world by the computer. By contrast, a human subject may express hesitation or uncertainty when reporting some poorly remembered information. This never happens to a computer. If an information is available, it is reported with absolute certainty as a known fact.
Humans can easily answer almost any question thanks to engaging emotional intelligence and employing experience they acquire in real life. A machine could not properly respond to figurative speech or idiomatic expressions, such as They took him to the cleaners. Also the common expression How are you doing? might be misunderstood by a computer. In all likelihood, the computer would reply, "How am I doing what?" To counter problems like these, common human expressions could be translated into machine operations that would simulate human responses. The computer might be programmed to automatically respond, "Thank you, I am doing fine", but would not understand the purpose of the exchange. The very act of "understanding" means something different in a machine than it means in man. When people understand an expression, they also experience (feel) its meaning based on previous encounters with the expression within a specific past context. This ability makes good lyrics, poetry, and prose appealing to people. A machine cannot "feel" the meaning of a word and cannot associate it with a previous contextual experience through internal electronic excitation. A machine cannot have an experience and cannot learn.
The denial of learning by machines sounds like heresy. There are research teams of Ph.D.'s who have built their careers around the concept that machines can learn. But from the human viewpoint, learning is only possible when an experience has a meaning. A machine does not care what happens. Data are just numbers, and a machine feels the same (that is nothing), no matter what data are processed. The words "guy" and "gentleman" appear the same to a computer. A computer has no ability to feel the emotional difference and choose the appropriate expression within the overall context. Furthermore, a computer does not know that it has learned something. A computer has no awareness of what information has been acquired and to what extent. Humans naturally employ their meta memory (awareness of what they know) and other mechanisms when they decide whether or not they know something and how well they know it. This awareness or lack thereof is not associated with conscious retrieval of the topic. Human beings can assess the extent of their knowledge without consciously retrieving the specifics of an information. By contrast, a computer has no ability to determine whether or not some data are available until the data are found. And since most computers only use a one-memory system, computers cannot confirm that the data are correct. The data are considered correct simply because they are found. If that mechanism alone were employed in human minds, then the memories of dreams or movies would be considered reality.
The above characterization does not mean that all forms of information acquisition are beyond the abilities of machines. If a machine were equipped with tactile sensors, the machine could move about and "learn" the layout of the surrounding environment by trial and error. The "learning" would be equivalent to spatial exploration in accordance with programmed instructions. The machine would likely be unable to distinguish between minor and gross mistakes, and unexpected obstacles or challenges could not be handled well. A machine lacking emotionally meaningful communication between its sensors and its experience stored in memory would not understand which maneuvers are permissible and which will lead to breakdown and self-destruction. Children learn these issues early in life as they explore their environment, but doing the same with machines is next to impossible. They would have to have operational limits specified in software to avoid doing things that might be dangerous. But real-life possibilities are countless. There would always be issues the designers never thought about. Similarly, the imposed restrictions might be too restrictive, and the machine might not attempt to perform functions that are doable. For example, touching a wall and hitting it at 10 mph might result in the same software reactions. Sure, the machine could be equipped with high-dynamic-range sensors that would sense various levels of impacts, but the machine would still feel nothing. Only the software would respond to the impacts with corrective measures. If the impacts occurred outside a sensor or destroyed the sensor, the machine would not register the impacts. In humans, damage to a sensor is detected by other sensors, and they inform the brain about the injury. These scenarios point out the crucial role of neural networks in the human body. The body and its sensors must be integrated with the brain to ensure the survival of the living organism. Although sensing of the status of the body of a machine might be technologically achievable one day, an insurmountable challenge exists in the "electronic brain," where the sensory information has to be interpreted.
It has been widely reported that the human brain has no sensors of pain, but this characterization is too simplistic. Although the brain cannot sense external pain applied directly to neural structures, the brain is needed to interpret signals of sensors. The human organism has perceptual neurons in the body and also in the brain. The two neuronal populations perceive, understand, and affect each other. Reactivity of the cerebral neurons allows you to experience pain, disgust, emotion, or consciousness. Clusters of brain neurons collect stimuli from the peripheral sensors and obtain a global picture about the state of the human body or the external world. In turn, the brain neurons affect the peripheral sensors, and act through them on other cells that constitute the human body. This activity is known as biofeedback. Both the peripheral sensors and the neurons in the brain can feel and experience the meaning of the processed information. This communication mode is not possible in machines. Machines can read the sensors, but the "electronic brain" feels nothing. The readings are processed in software (or by hardwired electronics) and have no impact on the physical state or "feeling" of the electronic circuits. The ability to feel and experience is what leads to sensible learning, intelligent choices, and free will in humans. Some human faculties have become so sensitive and specialized that the environment can be experienced through looks, language, sensory images, or mental imagination of various scenarios. Because of this dependency on living neurons and their responses, computational approaches cannot simulate the functions of the human brain.
The importance of biological interactions is apparent in daily life. Clinical work has shown that loss of meaningful contact with the external and internal physical worlds affects knowledge, intelligence, and relationship with oneself. Lack of feeling and perceptual knowledge results in the inability to distinguish the surrounding world from one's body. This deficit is known from patients with left-side paralysis that is accompanied by anosognosia. The subjects cannot recognize their left sides as parts of selves and produce confabulated answers about their bodies and physical abilities. Despite these deficits, the patients try to behave as if they were fully functional. A similar problem exists in patients with visual neglect. They can see in their neglected hemifields, but ignore the sensory messages. The cognitively unattended information has no significance in the brain and is ignored. Similarly, an intelligent machine that has no contact with its body might collect sensory information, but would fail to respond to the data, because they would have no practical meaning. A locomotive might be coming at the smart machine at 80 mph, and the machine would only "pay attention" to the exact position and speed of the locomotive. No idea would enter the "electronic mind" of the seemingly intelligent machine that it is 20 milliseconds from being obliterated. The clever machine would fail to get out of the way because the machine has no understanding of danger, pain, and death.
Another major problem with "intelligent machines" is how they are created. They are designed and produced by a centralized process. Everything the machine is comes from the design team. The software routines are programmed into the computer, but are not typically acquired by the machine in a process of learning. There are countless permutations that need to be accounted for to simulate real life, and countless interactive networks within the machine have to be implemented. The centralized design process can never be perfect in scope and quality. By contrast, a living organism develops in a distributed fashion. The conscious mind does not need to think about everything that is going on in the body and brain. Neurons interact locally and respond to problems locally. The brain only acts as a coordinator of strategies and higher functions. The brain does not tell neurons when to create synapses, store memories, or learn from experience. These details are spontaneously determined by the involved neurons by way of interaction with other neurons, sensors, or the social and physical environments. There are too many things going on in a living organism, and trying to account for them all by the programmer would be utterly foolish. Naturally, trying to design all such interactions into a machine is totally hopeless. Even people are consciously unaware of all their skills, knowledge, and experiences. Attempts to program such hidden information into machines are guaranteed to be unsuccessful.
Despite the striking imperfections of computers, scientists commonly use the expression "electronic brain" and imply that a thinking machine can be electronic. No thinking entity can ever be made of solid-state devices. The characteristics of living organisms hint that functional artificial intelligence can never be created in machines. A machine could be made to solve very complex numerical problems in the spirit of artificial intelligence, but the machine would have to be supervised by emotional intelligence. Lack of supervision would produce wrong or irrational solutions to problems, and would almost certainly lead to self-destruction of the "smart machine" or might cause harm to people. Solid-state machines that do not interact with the environment and do not feel its impact on the machines can never develop emotional intelligence. The only way to produce emotional intelligence (and also creative scholastic intelligence) is through highly responsive organic compounds. Very high organization and complex interactions of such compounds result in living cells that allow formation of more advanced multi-cell organisms. The characteristics of life indicate that the barrier between an inert machine and a smart life form can never be overcome. Machines are destined to remain dumb.
In recent years, neuroscientists have recognized the inherent limitations of electronic computers and have been trying to produce organic elements that could be usable in cybernetics. An example is the pursuit of an organic transistor. The transistor would supposedly have richer functional and logical repertoire than its silicon cousins have. The hope is that by applying such organic building blocks, intelligent machines can be produced. But will it work? The fundamental principle of intelligence demands that a biological element interact with the environment. This interaction is internally dependent on the element and is not a function of some externally applied instructions. What we call intelligence is mutual interactions of living cells. The physiology and biological needs of cells are the forces that produce interactions between neurons and lead to intelligence. An input from an external element that has no interaction with other cells and is not part of the whole system cannot induce intelligent, logical, beneficial, or desirable responses. This deficit can be seen in mental patients of all types. Giving such patients a sound advice is of no benefit if the patients are unable to evaluate the advice and recognize it as useful and beneficial. Naturally, organic computing elements are not living entities and are not comparable to neurons. The devices may be using organic compounds, but essentially behave like machines and only respond to programmed functions. Also in this case, scientists attempt to program intelligence through centralized design, but do not allow a living organism to become intelligent by virtue of its internal experiences, needs, and responses.
Since a computer has no feelings or needs, it has no necessity or ability to negotiate internal and external environmental effects. In turn, the lack of cognitive drives prevents a computer from seeking sensible measures to counter the environmental effects. As a result, a computer has no need to employ intelligence. In fact, the inert nature of computers makes them unfit to produce any intelligence whatsoever.
Many people have a problem imagining how the human mind could produce thought. The giant step from non-living chemical elements to intelligence seems inconceivable. In reality, the challenge is not that big. The people are forgetting that the human mind does not emerge from passive and largely inert chemicals. The giant leap has already been achieved in neurons. They respond to and interact with the environment. So, the human brain just employs large amounts of appropriately organized neurons to produce higher level of understanding of and interaction with the environment. The unfounded belief that the human brain (relative to simple life forms) employs some additional "mental magic" confuses many explorers of the brain and bothers them with artificially generated mysteries.
As the discussion hints, the unique properties of the brain cannot be duplicated by any other means but by life itself. A living cell with the ability to sense the environment and react to it is the basic building block of all higher organisms. Some cells build the body; some process information in real time; some store memories, and some facilitate metabolism. Although all cells are living individuals, they specialize and, similarly as social insects, fulfill unique functions within hybrid biological systems. Interactions of many neurons result in higher thought. There is no self-awareness at the level of a single neuron; the responses are "instinctive." Higher cognitive functions (emotion, self-awareness, and consciousness) only emerge as a product of interactions between huge numbers of neurons. These principles apply not only to life, but also to the existence of intelligence. An intelligent machine would have to be functionally structured the same way as the brain is. In reality, machines only mimic the overall behavioral manifestations of life, but lack the essential internal relationships that define life, feeling, emotion, intelligence, and consciousness.
 Jeff Prideaux. Comparison between Karl Pribram's "Holographic Brain Theory" and more conventional models of neuronal computation. Retrieved November 1, 2008 from http://www.acsa2000.net/bcngroup/jponkp/
 A Working Brain Model. Source: Complexity Digest 2007.46, 29-Nov-2007. Retrieved November 2, 2008 from http://www.comdig.org/index.php?id_issue=2007.46
 The Reptilian Brain and the Triune Brain Model (1/7). Retrieved November 2, 2008 from http://www.eruptingmind.com/reptilian-brain-triune-model/
 The Four Quadrant Model of the Brain, Ned Herrmann's Whole Brain Model. Retrieved November 1, 2008 from http://www.kheper.net/topics/intelligence/Herrmann.htm
 William Witherspoon (1999-2004). String Theory and the Human Mind. Retrieved November 2, 2008 from http://www.wwitherspoon.org/StringTheory.htm
 Bernhard Mitterauer & Kristen Kopp. The self-composing brain: Towards a glial–neuronal brain theory. Brain and Cognition, Volume 51, Issue 3, April 2003, Pages 357-367
 The Split Brain Experiments. Retrieved November 2, 2008 from http://nobelprize.org/educational_games/medicine/split-brain/background.html
 Francis Crick & Christof Koch. Consciousness and Neuroscience. Cerebral Cortex, 8:97-107, 1998
 Ignacio E. Ochoa Pacheco (1996). The Holographic Model of Brain Function and the Orgone Energy Fields Theory. Retrieved February 21, 2009 from http://www.orgone.org/articles/ax7ignc1.htm
 Yuste Rafael. Circuit Neuroscience: the road ahead. Frontiers in Neuroscience, July 2008, Volume 2, Issue 1, Pages 6-9.
|More information is accessible from the HOME page.|
|Unpublished work Š 2008-2017 Martin Dak. All rights reserved.|