Humans are extremely weak. We have no menacing teeth or sharp claws. We are clumsy: we cannot climb trees nearly as effectively as Chimpanzees, and we cannot run with anything approaching the speed of a Cheetah. We lack night vision. Our skin burns from sun exposure, and we suffer from allergic reactions to a plethora of natural elements.
Yet, despite all this, humans have managed to spread across all land masses on Earth. We may not have claws, but we have swords and knives, firearms, missiles, and thermonuclear bombs. We may be clumsy, but we have cars that can easily move us over 3 times as fast as a Cheetah can run; ships that can sail the seven seas; submarines that can carry us to the depths of the ocean; jets that can carry us faster than the speed of sound; and spacecraft that have taken us to the moon. We may lack night vision, but we have radar, sonar imaging, night-vision goggles, and telescopes than enable us to see into mind-bogglingly far reaches of the Universe. We may suffer from sun burns and allergies, but we have developed enormous, sophisticated cities, with towering skyscrapers, and other facilities that eliminate any need for us to venture outdoors unless it suits our fancy.
So how do we do it? What is that makes us so much different from the other forms of life on Earth? Intelligence - this is the key to the progress and survival of Homo Sapiens (“wise man”).
Intelligence, insofar as it exists in humans, is - at a fundamental level - the management of a model of the world, as a particular individual has observed it. This world model exists within our neocortex. Physically speaking, the hardware of these world models is made up of neurons, billions of them. They allow us to remember experiences, identify relationships between phenomena, and infer the results of our actions before they occur.
This world model obeys all the laws of physics, and we have absolutely no reason to believe there is any sort of supernatural component to it. The brain is perhaps the most complex system in the Universe - it certainly is the most complex that humans have yet been exposed to. So, it is not surprising that we have had so many misconceptions about it for so long. And even now, we have a very long way to go until we reach anything near a full understanding of how it works.
Our brains are plastic. This essentially means that preexisting connections between neurons can be changed, and new ones are able to form. When a human learns, they are reshaping their neural connections - thus updating their model of the world. Every individual’s world model is different, because the development of a world model is dependent upon the experiences of the particular individual that owns it.
There are two important attributes of world models:
Accuracy - the extent to which a particular world model is correct. Just because an individual updates their world model to reflect something that they think to be true, does not mean that they are correct. That is to say, individuals are perfectly capable of updating their world models in such a way as to no longer perfectly reflect reality.
Completeness - the degree to which a world model reflects all that may be known about the universe and reality. In general, the more an individual knows about the world, the more complete their model of the world is. Of course, it is impossible at present for the world models of humans to be entirely complete.
These two attributes are both very important, and generally share a direct relationship - but they are distinct concepts. An individual can have an accurate world model - one that contains no discrepancies with reality. But this does not necessarily mean that this individual is particularly knowledgeable. Suppose the individual simply has nothing of substance in their world model, as in the case of a newborn child. This brings to light another important point: World models ought not only be accurate, they should also be complete. That is to say, we wish for the content of individuals’ world models to reflect the truth of reality and we wish for their world models to contain as much information as possible about the universe.
One might ask why it matters if our world models are accurate. Who cares if we know the truth, do we not have the right to delude ourselves? Perhaps. But having accurate and complete world models allows us to act more powerfully. That is to say, if an individual has an accurate and complete world model, they are able to have a greater impact on reality than someone who does not. Our world models are attached to bodies. We have the capacity to affect and change the world (reality) with our bodies. But if our world models are not accurate or complete, then we are limited in our capacity to induce change. The accuracy and completeness of an individual’s world model has a direct correlation with the degree to which they can induce change.
For example, consider the obvious case of superstition vs science. Superstitious individuals - by definition - are far less likely to have accurate and complete world models than rational individuals who respect and participate in the process of science. For thousands of years, the majority of humans could be classified as superstitious. I do not wish to fault them because, by-and-large, they were raised that way. Their world models were polluted from the very beginning. Their world models were so corrupt that they were willing to torture themselves and others for no reason whatsoever (of course, they thought it was for a just cause - but we now know it was simply absurdity. Think: the Inquisition).
But then came the Enlightenment, and the Scientific Revolution. These movements brought about radical changes in human thought. We began to develop far more accurate and complete world models. Innumerable advances in the human condition were brought about by these changes. Suffice it to say, however, that the Enlightenment began sometime in the 17th or 18th century (1650-1700), and humans first walked upon the surface of the moon on July 20th, 1969. For untold thousands of years humans had barely been able to survive. We wallowed in superstition and ignorance. But after we developed and began to practice the process of science, the human condition advanced so rapidly that it took us only 300 years to reach the moon.
Science is said to be a “way of knowing”. But what does this really mean? I believe it refers to science as an algorithm that we use to manage the updating of our world models. Science encourages the acquisition of knowledge; the practice of free thought; the application of skepticism; and the support of evidence. This is why science is so powerful: 1) It allows us to identify and resolve discrepancies in our world model, and identify that which is true - thereby bringing our world models into closer alignment with reality. 2) It encourages us to share our model with others and collaborate with them - in effect, multiplying our computational power and effectivity.
Our world models are central to our intelligence, but we do not yet know on a deep level how the model works, or what the algorithms and mechanisms are that update and maintain it. Neuroscientists and Machine Learning researchers, as well as many others, are intent on understanding how the brain’s “software” works. The potential impacts of these endeavors are enormous. We will gain new insight into our strengths and fallibilities as a species. We will be better able to cure and prevent brain-related diseases. We will be able to enhance our own intelligence. And, perhaps most controversially, we will be able to improve our brain’s software, and use it to develop truly intelligent machines.
We need to figure out how our own world models are implemented in our own neural wetware, so that we can use similar techniques with silicon. We need to reverse-engineer the algorithms that manage our world models. How are they updated based on new experiences and exposure to novel ideas? How do we act upon them (behavior), and use them to induce change in the world? We need to gain a real, low-level, fundamental understanding of these elements, and how they work in our own brains. Once we have done so, we will be able to transfer similar, and improved, methods into silicon-based machines. These machines will then have intelligence similar, and undoubtedly superior, to our own.