How to Create a Mind: The Secret of Human Thought Revealed

by Ray Kurzweil

There are several reasons, however, why a skill or an area of knowledge that has been relearned using a new area of the neocortex to replace one that has been damaged will not necessarily be as good as the original. First, because it took an entire lifetime to learn and perfect a given skill, relearning it in another area of the neocortex will not immediately generate the same results. More important, that new area of the neocortex has not just been sitting around waiting as a standby for an injured region. It too has been carrying out vital functions, and will therefore be hesitant to give up its neocortical patterns to compensate for the damaged region. It can start by releasing some of the redundant copies of its patterns, but doing so will subtly degrade its existing skills and does not free up as much cortical space as the skills being relearned had used originally.

There is a great deal of redundancy in the patterns we learn, especially the important ones. The recognition of patterns (such as common objects and faces) uses the same mechanism as our memories, which are just patterns we have learned. They are also stored as sequences of patterns—they are basically stories. That mechanism is also used for learning and carrying out physical movement in the world. The redundancy of patterns is what enables us to recognize objects, people, and ideas even when they have variations and occur in different contexts. The size and size variability parameters also allow the neocortex to encode variation in magnitude against different dimensions (duration in the case of sound).

The techniques that we have evolved over the past several decades in the field of artificial intelligence to recognize and intelligently process real-world phenomena (such as human speech and written language) and to understand natural-language documents turn out to be mathematically similar to the model I have presented above. They are also examples of the PRTM. The AI field was not explicitly trying to copy the brain, but it nonetheless arrived at essentially equivalent techniques.

Although we experience the illusion of receiving high-resolution images from our eyes, what the optic nerve actually sends to the brain is just a series of outlines and clues about points of interest in our visual field. We then essentially hallucinate the world from cortical memories that interpret a series of movies with very low data rates that arrive in parallel channels.

“Even though we think we see the world so fully, what we are receiving is really just hints, edges in space and time,”

Everyone knows what attention is. It is the taking possession by the mind, in clear and vivid form, of one out of what seem several simultaneously possible objects or trains of thought. Focalization, concentration, of consciousness, are of its essence. It implies withdrawal from some things in order to deal effectively with others. —William James

Among its other functions, the thalamus is considered a gateway for preprocessed sensory information to enter the neocortex.

The most significant role of the thalamus, however, is its continual communication with the neocortex. The pattern recognizers in the neocortex send tentative results to the thalamus and receive responses principally using both excitatory and inhibitory reciprocal signals from layer VI of each recognizer. Keep in mind that these are not wireless messages, so that there needs to be an extraordinary amount of actual wiring (in the form of axons) running between all regions of the neocortex and the thalamus.

The Hippocampus Each brain hemisphere contains a hippocampus, a small region that looks like a sea horse tucked in the medial temporal lobe. Its primary function is to remember novel events. Since sensory information flows through the neocortex, it is up to the neocortex to determine that an experience is novel in order to present it to the hippocampus. It does so either by failing to recognize a particular set of features (for example, a new face) or by realizing that an otherwise familiar situation now has unique attributes (such as your spouse’s wearing a fake mustache).

memories in the hippocampus are also stored as lower-level patterns that were earlier recognized and stored in the neocortex.

The capacity of the hippocampus is limited, so its memory is short-term.

We need, therefore, a hippocampus in order to learn new memories and skills

Someone with damage to both copies of her hippocampus will retain her existing memories but will not be able to form new ones.

The hippocampus is one of the first regions damaged by Alzheimer’s,

The Cerebellum There are two approaches you can use to catch a fly ball. You could solve the complex simultaneous differential equations

This is not the approach that your brain adopts. It basically simplifies the problem by collapsing a lot of equations into a simple trend model, considering the trends of where the ball appears to be in your field of vision and how quickly it is moving within it. It does the same thing with your hand,

is called basis functions, and they are carried out by the cerebellum,

The cerebellum is an old-brain region that once controlled virtually all hominid movements.

Fear is the main source of superstition, and one of the main sources of cruelty. To conquer fear is the beginning of wisdom. —Bertrand Russell

The earliest brains evolved pleasure and fear systems that rewarded the fulfillment of these fundamental needs along with basic behaviors that facilitated them. As environments and competing species gradually changed, biological evolution made corresponding alterations. With the advent of hierarchical thinking, the satisfaction of critical drives became more complex, as it was now subject to the vast complex of ideas within ideas. But despite its considerable modulation by the neocortex, the old brain is still alive and well and still motivating us with pleasure and fear.

Pleasure is also regulated by chemicals such as dopamine and serotonin. It is beyond the scope of this book to discuss these systems in detail, but it is important to recognize that we have inherited these mechanisms from our premammalian cousins. It is the job of our neocortex to enable us to be the master of pleasure and fear and not their slave.

Sometimes we, like the rats who died overstimulating their nucleus accumbens, use a shortcut to achieve these bursts of pleasure, which is not always a good idea.

Gambling, for example, can release dopamine, at least when you win, but this is dependent on its inherent lack of predictability. Gambling may work for the purpose of releasing dopamine for a while, but given that the odds are intentionally stacked against you (otherwise the business model of a casino wouldn’t work), it can become ruinous as a regular strategy.

A particular genetic mutation of the dopamine-receptor D2 gene causes especially strong feelings of pleasure from initial experiences with addictive substances and behaviors, but as is well known (but not always well heeded), the ability of these su...

Serotonin is a neurotransmitter that plays a major role in the regulation of mood. In higher levels it is associated with feelings of well-being and contentment. Serotonin has other functions, including modulating synaptic strength, appetite, sleep, sexual desire, and digestion. Antidepression drugs such as selective serotonin reuptake inhibitors (which tend to increase serotonin levels available to receptors) tend to have far-reaching effects, not all of them desirable (such as suppressing libido).

The amygdala is also part of the old brain and is involved in processing a number of types of emotional responses, the most notable of which is fear. In premammalian animals, certain preprogrammed stimuli representing danger feed directly into the amygdala, which in turn triggers the “fight or flight” mechanism. In humans the amygdala now depends on perceptions of danger to be transmitted by the neocortex.

our emotional experiences take place in both the old and the new brains. Thinking takes place in the new brain (the neocortex), but feeling takes place in both.

if it is just human cognitive intelligence that we are after, the neocortex is sufficient.

There is a continual struggle in the human brain as to whether the old or the new brain is in charge. The old brain tries to set the agenda with its control of pleasure and fear experiences, whereas the new brain is continually trying to understand the relatively primitive algorithms of the old brain and seeking to manipulate it to its own agenda.

It is important to point out that these cells are not doing rational problem solving, which is why we don’t have rational control over our responses to music or over falling in love. The rest of the brain is heavily engaged, however, in trying to make sense of our mysterious high-level emotions.

Interestingly, spindle cells do not exist in newborn humans but begin to appear only at around the age of four months and increase significantly in number from ages one to three. Children’s ability to deal with moral issues and perceive such higher-level emotions as love develop during this same period.

Neocortical abilities—for example, the ability of the neocortex to master the signals of fear that the amygdala generates (when presented with disapproval)—play a significant role, as do attributes such as confidence, organizational skills, and the ability to influence others. A very important skill I noted earlier is the courage to pursue ideas that go against the grain of orthodoxy. Invariably, people we regard as geniuses pursued their own mental experiments in ways that were not initially understood or appreciated by their peers.

One approach to expand the available neocortex is through the collaboration of multiple humans. This is accomplished routinely via the communication between people gathered in a problem-solving community.

The next step, of course, will be to expand the neocortex itself with its nonbiological equivalent. This will be our ultimate act of creativity: to create the capability of being creative. A nonbiological neocortex will ultimately be faster and could rapidly search for the kinds of metaphors that inspired Darwin and Einstein. It could systematically explore all of the overlapping boundaries between our exponentially expanding frontiers of knowledge.

this additional intelligence will essentially reside in the cloud (the exponentially expanding network of computers that we connect to through online communication), where most of our machine intelligence is now stored. When you use a search engine, recognize speech from your phone, consult a virtual assistant such as Siri, or use your phone to translate a sign into another language, the intelligence is not in the device itself but in the cloud.

We have already outsourced much of our personal, social, historical, and cultural memory to the cloud, and we will ultimately do the same thing with our hierarchical thinking.

we are now able to identify the biochemical changes that occur when someone falls in love. Dopamine is released, producing feelings of happiness and delight. Norepinephrine levels soar, which lead to a racing heart and overall feelings of exhilaration. These chemicals, along with phenylethylamine, produce elation, high energy levels, focused attention, loss of appetite, and a general craving for the object of one’s desire.

Studies of ecstatic religious experiences also show the same physical phenomena; it can be said that the person having such an experience is falling in love with God or whatever spiritual connection on which they are focused.

From an evolutionary perspective, love itself exists to meet the needs of the neocortex. If we didn’t have a neocortex, then lust would be quite sufficient to guarantee reproduction. The ecstatic instigation of love leads to attachment and mature love, and results in a lasting bond. This in turn is designed to provide at least the possibility of a stable environment for children while their own neocortices undergo the critical learning needed to become responsible and capable adults.

We have already largely succeeded in liberating sex from its biological function, in that we can have babies without sex and we can certainly have sex without babies. The vast majority of sex takes place for its sensual and relational purposes. And we routinely fall in love for purposes other than raising children.

The neocortex is biology’s greatest creation. In turn, it is the poems about love—and all of our other creations—that represent the greatest inventions of our neocortex.

The evolution of animal behavior does constitute a learning process, but it is learning by the species, not by the individual, and the fruits of this learning process are encoded in DNA. To appreciate the significance of the evolution of the neocortex, consider that it greatly sped up the process of learning (hierarchical knowledge) from thousands of years to months (or less). Even if millions of animals in a particular mammalian species failed to solve a problem (requiring a hierarchy of steps), it required only one to accidentally stumble upon a solution. That new method would then be copied and spread exponentially through the population.

When we augment our own neocortex with a synthetic version, we won’t have to worry about how much additional neocortex can physically fit into our bodies and brains, as most of it will be in the cloud, like most of the computing we use today. I estimated earlier that we have on the order of 300 million pattern recognizers in our biological neocortex. That’s as much as could be squeezed into our skulls even with the evolutionary innovation of a large forehead and with the neocortex taking about 80 percent of the available space. As soon as we start thinking in the cloud, there will be no natural limits—we will be able to use billions or trillions of pattern recognizers, basically whatever we need, and whatever the law of accelerating returns can provide at each point in time.

we will be able to back up the digital portion of our intelligence. As we have seen, it is not just a metaphor to state that there is information contained in our neocortex, and it is frightening to contemplate that none of this information is backed up today. There is, of course, one way in which we do back up some of the information in our brains—by writing it down. The ability to transfer at least some of our thinking to a medium that can outlast our biological bodies was a huge step forward, but a great deal of data in our brains continues to remain vulnerable.

If the Blue Brain Project brain is to “speak and have an intelligence and behave very much as a human does,” which is how Markram described his goal in a BBC interview in 2009, then it will need to have sufficient content in its simulated neocortex to perform those tasks. 6 As anyone who has tried to hold a conversation with a newborn can attest, there is a lot of learning that must be achieved before this is feasible.

There are two obvious ways this can be done in a simulated brain such as Blue Brain. One would be to have the brain learn this content the way a human brain does. It can start out like a newborn human baby with an innate capacity for acquiring hierarchical knowledge and with certain transformations preprogrammed in its sensory preprocessing regions. But the learning that takes place between a biological infant and a human person who can hold a conversation would need to occur in a comparable manner in nonbiological learning. The problem with that approach is that a brain that is being simulated at the level of detail anticipated for Blue Brain is not expected to run in real time until at least the early 2020s. Even running in real time would be too slow unless the researchers are prepared to wait a decade or two to reach intellectual parity with an adult human, although real-time performance will get steadily faster as computers continue to grow in price/performance. The other approach is to take one or more biological human brains that have already gained sufficient knowledge to converse in meaningful language and to otherwise behave in a mature manner and copy their neocortical patterns into the simulated brain. The problem with this method is that it requires a noninvasive and nondestructive scanning technology of sufficient spatial and temporal resolution and speed to perform such a task quickly and completely. I would not expect such an “uploading” technology to be av...

There is a third approach,

simplify molecular models by creating functional equivalents at different levels of specificity, ranging from my own functional algorithmic method (as described in this book) to simulations that are closer to full molecular simulations. The speed of learning can thereby be increased by a factor of hundreds or thousands depending on the degree of simplification used. An educational program can be devised for the simulated brain (using the functional model) that it can learn relatively quickly. Then the full molecular simulation can be substituted for the ...

A neural net starts out ignorant; its teacher—which may be a human, a computer program, or perhaps another, more mature neural net that has already learned its lessons—rewards the student neural net when it generates the correct output and punishes it when it does not. This feedback is in turn used by the student neural net to adjust the strength of each interneuronal connection. Connections that are consistent with the correct answer are made stronger. Those that advocate a wrong answer are weakened.

We could train our device to learn the patterns for a particular person using a moderate-sized vocabulary, measured in thousands of words. When we attempted to recognize tens of thousands of words, handle multiple speakers, and allow fully continuous speech (that is, speech with no pauses between words), we ran into the invariance problem. Different people enunciated the same phoneme differently—for example, one person’s “e” phoneme may sound like someone else’s “ah.” Even the same person was inconsistent in the way she spoke a particular phoneme. The pattern of a phoneme was often affected by other phonemes nearby. Many phonemes were left out completely. The pronunciation of words (that is, how phonemes are strung together to form words) was also highly variable and dependent on context. The linguistic rules we had programmed were breaking down and could not keep up with the extreme variability of spoken language. It became clear to me at the time that the essence of human pattern and conceptual recognition was based on hierarchies. This is certainly apparent for human language, which constitutes an elaborate hierarchy of structures. But what is the element at the base of the structures? That was the first question I considered as I looked for ways to automatically recognize fully normal human speech.