One of the many things we do when we are alive is create simplified explanations for very complex things. We create structure out of chaos, and that structure we build will never be completely perfect, because a true solution comes from such a deep understanding and level of analyzation that most humans simply are not capable of. We discovered mathematics, a structure and model of how things interrelate with each other in their most fundamental, numerical form. We created science for the purposes of exploration, understanding, and scientific findings fundamentally arises from curiosity about the mechanisms of the universe and nature. We created philosophical and religious principles though which we express our gift of spirituality and morality. We are given chaos in every aspect of life, and we deal with this chaos by understanding the order within the chaos, and building internal models to direct us to order and away from chaos. We are directed towards sources of understanding, because understanding is a formation of order from the chaos.
When we study neuroscience we attempt to understand how the brain works at some of its most fundamental levels. The deepest, most fundamental level of neuroscience is the level on which neurons function. We know, at least fundamentally, how a single neuron receives inputs from neurotransmitters, and outputs neurotransmitters given an input. It is still very unknown the specific of how billions of neurons fit together to work in the perfect way that they do, but we as scientists will find a way to uncover that mystery. Some of the levels above the most fundamental system include studying respective systems and their functionality inside of the brain, such as the limbic system with emotions. Fear is one of the most prevalent emotions in our everyday life, fear can control us, and drive us.
One of the way that we derive order from the unknown structure in the brain is to figure out its psychological mechanisms that drive our emotions. You can think about your brain in three distinct levels, at the bottom there is the millions of years old, ancient lizard brain, comparable to the Freudian 'Id'. This section of your psyche and even physical brain, is conceptually on the bottom of the other layers, it is responsible for your heart beating, for breathing, and for all of the other unconscious regulatory behaviors. It is called a lizard brain, because it is not to be thought of as too much different from a lizard. Lizards do not have emotions, and they do not have complex frontal cortex behavior. On top of the lizard brain level, we have the emotional level, we call this the limbic system. We start to see emotional behavior in animals that exist in groups and hunt in groups. There is a lot of limbic behavior that goes into chasing a prey with a tribe of your own. The limbic system is an evolutionary adaptation that binds groups together with emotions, and it gives them internal feedback from their environment. Conceptually, on top of all of these layers, is the newest evolved system, primarily found in primates, the cortex. Neuroendocrinology is the study of how these systems interact with each other.
You can think about the limbic system talking to the lizard brain, you feel some type of fear, and your heart starts beating fast and you start breathing fast. It can also go in the opposite way, and create some negative feedback loop. Your heart starts beating faster naturally, then you start to freak out and have fear, then it just goes back to your heart beating faster. The cortex is capable of abstract thought, and it is an expert in empathy and sympathy. Imagine sitting in a horror movie, your cortex imagines your are in that place in the movie, and you start to sweat and your heart starts beating. That would be your cortex talking all the way down to the primordial, lizard brain. We use bio-feedback loops to solve our problems, for example, if we calm our lizard brain we can decrease the risk of a stroke or heart attack.
How does this relate to programming at all? When we are trying to go about solving artificial general intelligence, one way to think about it is mimicking a human brain. Understanding the human brain goes beyond loosely mimicking the structure of neurons, understanding the human brain goes into understanding all of the aspect of life. We need to understand how we perceive the world, how we process language, and how the evolution behind us has influenced us today. We are perfect machines, perfected over billions of years of evolution. It is infeasible to mimic the product of so much time with just a brute force approach of adding more and more parameters to a neural net.