Why young minds often come up with the paradigm shifting ideas

I am going to be making some statements without explaining too much. The statements are explained in some of the other essays I have written, so I am not going to spend too much time explaining from first principles. I'll eventually add the hyperlinks.

Entropy is a fundamental principle of the universe and the only known principle in physics that can distinguish between the past and the present. The third law of thermodynamics states that the entropy of any closed system can only increase, never decrease. This simple rule has embedded in it the mechanism for the emergence of all the complexity we experience, including the complexity of intelligent life, arguably the most fantastic byproduct of the universe. 

Mathematically, the entropy of a particular configuration is a measure of probability, specifically of that configuration, over all possible configurations that make up a macrostate. High entropy means that the probability of occurrence of that configuration is high, while a low entropy means that the probability of occurrence is low. For example, tossing 100 coins simultaneously, the probability (and thus the entropy) of the configuration 50-Heads, 50-Tails is highest, whereas that of 100-Heads is the lowest. Low entropy sequences is the same principle applied to sequences.

(credit: Hyperphysics)

Matter, life, fuel, chemicals, energy and almost everything we interact with spans a wide range of probabilities, but the most interesting, valuable and critical things are inherently low entropy. The wood you burn to generate a fire takes an ordered configuration of carbon molecules mixed with other atoms and loses most of its structure, leaving behind ash. The food we eat is low entropy configurations. 

Entropy is so fundamental to the universe that the brain as an organ evolved to predict entropy. We don't need a clear definition of life; observing something at our scale is enough to distinguish between animate and inanimate objects. Thus the brain, in the long evolutionary run, acquired the ability to distinguish the low entropy events and sequences and store a mental representation of that as memory. 

Due to the complexity of our brain and the number of neurons we have been endowed with as a species, there are shared abilities common across our species. These include the ability to use tools, learn and speak language, make sense of what we are looking at, understand speech, recognize faces, and so many more. These subconscious abilities are programmed through billions of years of evolution and are a repository of fundamental low-entropy events and sequences. G-factor or intelligence is a measure of the brain's ability to sift through low entropy memory space and see similarities (what we call pattern recognition). Our intelligence or g-factor is the ability to build up our repository with new low-entropy sequences by learning and understanding new concepts. Learning and understanding are associating a new event that the brain calculated as low entropy to others in the repository.

Thus g-factor is the ability to take our repository of low-entropy memory space and apply it to a new event we haven't encountered yet. The repository helps predict entropy and see the similarity to something in our memory. 

This brings us to fitting. In artificial intelligence, overfitting is when the neural network becomes so good at predicting the idiosyncrasies of the training data that its performance on the training set is very high. However, show the same network an example it has not seen, and it paradoxically struggles. Thus the network is unable to generalize. Underfitting is a better place to be in, where the network's performance is reasonably high and thus has learned the general characteristics in the training data. Thus it can extract the same characteristics in a sample not encountered and offer a better prediction. 

Under and overfitting find an analogy in knowledge work like science and entrepreneurship. Many groundbreaking ideas come from young minds who have a high g-factor but are learning that subject for the first time and hence aren't bogged by the biases and idiosyncracies of the past. They come at it with a "fresh pair of eyes." I think this is analogous to older professors overfitting the data because of how much time they have spent, and the younger minds underfitting and thus can generalize and pick the most interesting threads to pull that may unravel the problem. 

Notice I said young minds and not youngsters. By this, I mean those who are learning a subject matter anew and constantly adding to their repository. As long as you are constantly adding to your repository, you are young in your mind. When you say most of what you need to know you know, you start tending towards overfitting. The brain is so complex and magnificent that even with an overfit you can still do amazing work, but the paradigm-shifting ideas will likely stay out of reach. 

I have been teaching myself reinforcement learning and have made some interesting connections that I want to share. In reinforcement learning (RL), you use games as a substrate to teach an AI how to learn the optimal strategy to win. You craft an algorithm that the RL agent, given enough computing power, can theoretically converge to the optimal strategy to play the game. 

This is the algorithm that was used by Deepmind to train AlphaGo beat a human player at the game of Go. A truly marvelous achievement. 

The way you do this is to teach an agent the value of the specific state the game can take. An RL algorithm has a few essential components, an environment, a state, a transition function, a value function, and a reward function. 

Essential concepts of a Reinforcement Learning Algorithm

• In the context of an RL agent, this is the game you are playing. So if an agent has been trained on the game of chess, then the board, the pieces, the rules to their movement is the environment. Essentially the game is where the agent gets to play and receive feedback on the moves they make. 
• Since RL algorithms are so deeply inspired by biological agents such as humans, the analog to an environment for RL agents in the real world. The universe and the earth we inhabit are the environments where we act in, and give us feedback on our actions. 

State 

• Each distinct configuration of the environment is a state. So in the case of the chess board, a specific position of the pieces is a state. The total number of possible states in chess is humongous, but these also include impossible states given the game. (e.g.: while a state where the king is in the front row and a pawn is behind it in the zeroth row is technically possible, but impossible given how the game starts and is played.)
• Even considering all possible legal states of a chess board, the number is humongous - it is on the order of 1E45.

Action

• An action is the set of allowed moves that the agent can take that transitions the environment from one state to another. For example, moving the pawn one step ahead transitions the environment from one state to another. 

Transition function 

• This is a learned function that takes as an input the current state, and an action and outputs the resulting state of the environment i.e., how to transition the environment from one state to another given an action. (e.g.: moving the pawn by one step forward is the action, the current configuration is the input state, and the resulting configuration is the output)
• Since many moves are possible, the transition function may have many entries for a particular state, but for the combination of state and action, there is only one possible mapping.

Value function 

• This is the value of every state that has been encountered. The value is learned through the algorithm, and this entry is how the RL agent decides which action to take, given their current state. If two different actions are possible, transitioning into two different resulting states, and the value of the first state is higher than the second, then the agent will choose the first action. 
• The concept of value is an abstract measure of how valuable a state is in attaining the final goal. In chess, the goal is to win the game, and not all prior states are equal in value. The ones where you control more of the board and more of your pieces are on the board is a better state to be in. In comparison, having your opponent control most of the board and have more pieces in play than you is a worse state to be in. Thus these intermediary states have differing abstract values in the pursuit of the win. It is these values that the RL algorithm learns. 

Reward function

• The reward for the action you take. In its purest definition, the only reward of playing a game of chess is if you win. The complexity is in figuring out intermediate rewards, otherwise the agent would be randomly moving through the state space until it reaches a win. This is akin to a group of monkeys randomly banging on a typewriter for perpetuity, and ultimately one of the random combinations would be a coherent sentence and one would be Shakespeare's works. 

The explosion of the state space

Now we have a sense of the essential components of an RL algorithm. As we start with simple games, we can encounter every state and create an entry in the "value function" table. For example, in the game of tic-tac-toe, there are 3^9 (or 19,683 possibilities) which a computer can easily work with. Note that this number includes illegal states, so the actual number is smaller. A minor upgrade to a game like (Connect-4)[https://en.wikipedia.org/wiki/Connect_Four] where you drop a coin and aim to make the longest sequence of 4 coins, the number of states explodes to 3^42 (or 1.09418989E20)! The total bytes of information humanity has produced so far in totality is estimated to be 1E22 bytes. This is comparable to the state space of a simple children's game! It is a wonder we are able to play it at all. 

If you start conjuring up state spaces for games like chess and Go, the number of possible states is truly mind-boggling. Go, for example, has 1E170 (1 followed by 170 zeros!) possible states. The total number of atoms in the universe is estimated to be 1E80. So you can appreciate the magnitude of the size of the state space of Go. Yet we humans play it, and the experts play it profoundly well. 

Ultimately, the reason that RL works is that it is modeled after human intelligence. When we are born, we are not given 10,000 labeled images of a car; we see a car, we interact with it, which is feedback from the environment and store quite quickly the concept of a car. This is especially more visible in our actions as adults. If we decide to go to the supermarket, there are theoretically infinite number of microsteps we take to reach our goal. If a particular road is blocked, we find a way around it. Even if this means passing through an unfamiliar part of the environment, we do it since our focus is on reaching the supermarket. A game is a "toy environment" created to exercise the same faculties. 

When we play a game, we do not map the entire state space and calculate the best move. In some intuitive way, we know which of the intermediate states are valuable and which positions are advantageous even if we haven't seen or experienced those states before. That is how we can comprehend a game like Go with its mind-boggling complexity and still play creatively. 

Beauty and state space reduction 

I posit that the abstract word we use, "beauty," encapsulates the value of state space. This word may go down as the most complex word to mathematically define because if we can, then we can program it into an AI. The AI then may be able to exercise creativity of the sort that is, even now (with all the advance of AI), firmly in the human realm of possibility. Mathematicians talk about elegance and beauty as a precept for the correctness of new mathematical work. The greatest equations are often transfixingly simple looking. Scientists who discovered new paradigms like how atoms are structured, how electricity flows, and the influence of DNA on life, know they are right even if they don't have the proof yet. The ideas often seep into their minds years before they rigorously prove their idea. Einstein, as an example, called the first time he thought of the equivalence of frames of reference for acceleration as his happiest thought. It took him many years to compose it into a Theory of Relativity. 

If a simple-looking game like Go has so many possible states, imagine the game of life. The number of possible states is truly and completely infinite, in the most infinite sense of the word infinity. We still manage to move forward and, most interestingly, discover new science. We are able to because we are guided by our innate sense of beauty that has guided us with an intuitive sense of valuing states. When you learn new science, and you truly understand, you experience that sense of beauty. Over time you hone that sense and start recognizing new ideas that have the potential of greatness. 

An AI does not yet have that ability to distinguish beauty or feel that trembling sense of awe when you understand how sunlight is converted to glucose in chlorophyll, and how that is responsible for the complex life on earth. It even applies to other humans; we see beauty in people (both physical and beyond), it is that same sense at play. The truly deep experiencers of life have figured out a way to recognize this sense of beauty at levels deeper than the our automatic reflex sense. 

We all have the mechanism to recognize beauty, it is on us to put it into practice. 

Why an over-reliance on system-thinking may be holding you back

Table of Contents: 

• An analogy between an over-reliance on genetic gifts: brute-strength and logic-based intelligence 
• Types of thinking - Accumulatory vs Exception based 
• The bias in choosing the pattern 
• Escaping the bias of low-depth pattern repositories 
• Becoming multidimensional thinkers

Notes: - I use the word pattern but the meaning is interchangeable with first-principles, first principles are a collection of axiomatic patterns, patterns we internally do not question the veracity of. The more you have of these, the more complex patterns you are able to build by using the first-principles like lego building blocks of logic.

An analogy between an over-reliance on genetic gifts: brute-strength and logic-based intelligence

We live in a period of unprecedented peace. After all, until about 80 years ago mankind was a warring species. This becomes even more true if you go further back. Physical security was constantly at risk, and hence, the male of the society may have come to be prized and preferred for their physical abilities. This was not the only reason but in my view an important one. In these times, the highest achievements were associated with the physical capabilities of the body. Thanks to their genetics, people who were born with brute strength were able to easily overpower the average and the weaker. These people, you can imagine, did not have to work very hard to earn their keep. In any war, an excess of 90% of the soldiers are within two standard deviations from average. Those gifted with abilities three standard deviations out could conceivably fight many battles before coming upon someone who posed a threat. If this opponent was similarly gifted, then it was a battle of innate strengths; the more gifted one wins.

However, there was a different kind of opponent that they could encounter. The ones who went beyond their strength (or lack thereof) and disciplined themselves into becoming the finest warriors they could. From a young age, these individuals honed their physical capabilities to their limit, the difference being that they don’t rely on a single overpowering ability. Put simply, they don’t rely just on their strength. They develop the ability to strike at vital points so that they are most efficiently able to immobilize a person. Even the strongest person is not much of a threat if their knee cap or legs are taken out. They practice their swordsmanship so they can parry a blow before it comes so that they can see an opening in an attack that the attacker themself isn’t even aware of. They are able to tell which way a sword is going to swing based on the grip, the leaning of the body weight and the movement of the eyes. These are subtle clues that operate at the most granular level of operation. You wouldn’t think to consider such details if you are quite strong, you never needed to rely on these subtleties. Your strength was enough.

Fast forward to the present world, pure physical strength is not valued as much by society because of unprecedented peace. Now we value mental prowess, what we generically refer to as intelligence. Howard Gardner posits the existence of at least 8 types of intelligence that go beyond the purely cognitive ones we like to default to. These include kinesthetic, clerical, spatial, musical and logical (to name a few). Let us focus on logical intelligence, which is associated with a more common-sense meaning of the word intelligence. A particular aspect of this that is favored in technical jobs, academia, computer science applications, and scientific research is “systems thinking”.

Types of thinking - Accumulatory vs Exception based

Thinking in systems is a powerful ability. It is the ability to take a set of facts and glean the underlying pattern that connects those facts. You could say that most people are system thinkers by that definition, and that is generally true. However, the difference is in the edges. There are two types of thinking that I will lay out, one is called the accumulatory and the other is called the exceptional. In accumulatory thinking, the brain takes a set of facts and is constantly coming up with a hypothesis of the underlying pattern and adopting that as a general rule until contrary evidence presents itself. When contrary evidence presents itself, the pattern is updated to account for the new fact and updated throughout.

For example, seeing a few chairs you are able to generate a hypothesis pattern that chairs usually have 4 legs, a base, and a backrest. This pattern holds until a chair with 3 legs shows itself. You update the general pattern of the number of legs and leave being a stable, supported system with a base and backrest unchanged. Then you come across an office chair with wheels, and now you again update the rule. Then you come across a stool, and you realize this can also function as a chair, but it doesn’t have a backrest, so you generalize the definition to something stable you can sit on. Then you come across a beanbag, and that cannot be called a stable system. So you further refine your pattern of a chair. This process continues, indefinitely. However, with each change, the probability of a subsequent change decreases.

Since most of us do this without even trying, we can see that this is a general skill humans have. The difference is in our ability to use this skill in more complex domains. Science is the process of extracting these underlying patterns by running experiments and using those as the facts to build a hypothesis of a pattern on. Since we know not all of us can be scientists, there must be another way of thinking.

The exceptional way of thinking is one where the most important characteristic is not the pattern but the exception. If a pattern can explain 80% of the actual happening (which is a pretty good explanatory power) the pattern is still not accepted and internalized because of the 20% that it cannot. An example of this is in the work of Noam Chomsky, who has extensively studied language out of his own curiosity and noticed the pattern of similarity in the mechanics of language across cultures, geographies, and time; and posited the general pattern of a Universal Grammar. This is an appreciation of the fact that human brains seem to have a common underlying structure that seems to hold a universal grammar system that enables us to learn something as complex as language so easily when we are young. While he calls it universal, people have found remote tribes in the Amazon that seem to have developed a simple language system that deviates from the Universal Grammar rules that Chomsky laid out. In essence, a vast majority of languages seem to fit the pattern, but exception based thinkers argue that we cannot give it as much importance because of the few deviations observed from it. Here the important thing is not the pattern, but rather the exception. That is what I mean by exception thinkers.

Exception thinkers do very well in operations, supply chain, logistics, crime-analysis, and so on where the most important things are the exceptions, while accumulators make good scientists, entrepreneurs and politicians.

The bias in choosing the pattern

Hence system-thinkers tend to be accumulators. As accumulators the skill is in choosing the pattern. Recognizing novel patterns on the fly is extremely hard, especially in adulthood. When we are children we engage in some of the most sophisticated pattern recognition and accumulations that we will ever do in our lives. The reason is that a baby’s brain is a hyperconnected collection of neurons. This means that our brain is never more interconnected in its neurons than it is when a brain is in early development. Childhood and puberty is the sequential pruning of these connections, deciding which connections to keep and which ones to discard. This hyperconnected brain is an incredibly sophisticated pattern recognizer. No wonder something as complicated as language is understood so easily by the brain. This is also when we play games, form social connections, and learn love from our family. So many of our abstract patterns are learned at this stage. Our interests, proclivities, biases, and experiences shape this initial set of patterns we build a repository of.

However, no matter how intelligent we may be, we are a reflection of the diversity of the universe we experience. As children, we experience only a small cross-section of the universe from a relatively safe vantage point (care of our parents). It is rarely representative of the realities of the world. However, we don’t know that, and to us the patterns we build seem fairly robust. So when a system-thinker identifies a pattern that explains the facts they witness, they are drawing from this repository no matter the depth of the repository.

This reflects the analogy of the physically gifted brute strength of a soldier. As somebody who has relied on their genetic predisposition and ability to system-think, and never really developed it, they too rely on their given strength, and rarely invest the time to develop that skill. If you personify the system-thinking ability, it develops a false sense of importance from being always relied upon. This is not unlike a merchant who is a middleman that passes the goods to the appropriate recipient, doesn’t add much value but since all trade passes through them they grow wealthy and arrogant. They do not allow the development of other businesses that may pose even a remote threat to their hegemony. In this way, they suppress innovation, unless it is of the type that helps them grow their own influence and become even wealthier. The arrogance develops into selfishness. If the system-think is not able to system-think, it questions the data, even the very problem it is given to solve. If I cannot see the larger pattern here at a first glance, then maybe the problem is wrong or unsolvable, or at minimum incomplete.

Escaping the bias of low-depth pattern repositories

Is there a way out then? Let’s recap the weakness imminent in always relying on system thinking. We know that system thinking allows you to quickly glean the pattern that explains the data. The individual is less worried about the relevance of the pattern, rather that they were able to see a pattern. However, these patterns are typically drawn from a pattern repository. This repository can also be thought of as a collection of first-principles. If you have very few patterns to pull from, then your system thinking is going to go around in circles or choose the same one repeatedly. What you then need to do is add to this repository. But isn't that hard in adulthood?

Yes, it is. There is still a way out, and it is easier because we live in an information age. If system thinking is the input of facts or experiences and the output of a pattern (or series of patterns with a hierarchy of their own), then how about changing those two variables. Philosophy and logic is the study of the art of developing patterns that have been collectively honed in the minds of the extraordinary humans that came before us. Study of philosophy and logic helps you avoid making false equivalences, see a pattern’s overfit or underfit, and look beyond your personal bias in building up patterns. The other way is the study of science, where people like Galileo, Newton, Darwin, Turing, Maxwell, Einstein, Planck, and Dirac, to name a few, spent their lives dedicated to the study of the universe and gleaned some of the most profound patterns. If philosophy is the meta-skill to aid pattern extraction, then science is an incredible repository of patterns gained from the study of the universe.

The universe is THE repository of applied patterns, where certain patterns find resonance and are elevated at different scales of reality. As an example, gravity is a pattern of particles gaining power in accumulations, the larger the accumulation the larger the pull. This resonates at the level of atoms, stars, and planets, but also at the level of an organization in human societies. The bigger the organization, bigger the pull.

Some of the simplest patterns are incredibly far-reaching, as the emergence of evolution and descent into chaos of entropy. Quite likely evolution and entropy were the basis of yin and yang. The human body survives when homeostasis is maintained, balancing the calming (sympathetic nervous system) and the excitatory (parasympathetic nervous system). In our brain too, we have excitatory and inhibitory neurotransmitters. Evolution and entropy are far-reaching in their implication. These were but a few systems at different levels of scale that manifest the applied patterns.

Another way to improve your system-thinking is (ironically) by suppressing it, and allowing other skills to develop. The personification of system-thinking is quite apprehensive of this move. It does not like being relegated to a non-primary position where it is not the lead character running the show. So it protests and tries to create pandemonium. When the slightest hints of difficulty arise from deliberately focusing on other skills, system-thinking protests the loudest and screams - see, you need me, look at you unable to find the solution that I so obviously can see. Put me back in charge, and I’ll show you how it’s done. This is the trap laid by the loudest skill in your repository, the one that enjoyed the spoils of being the most experienced and adored skill. So it’s easy to fall back into that trap and seek the comfort of knowing rather than go through the discomfort of deliberately thinking.

Becoming multidimensional thinkers

The beauty is that system-thinking is a meta-skill, like how philosophy and logic are meta-skills, specifically the art of recognizing patterns. A meta-skill does not grow stronger by itself, it is only as strong as the number of hard skills that it has access to. So when you sharpen the other skills, your system-thinking grows stronger. It also gains hubris and realizes that it doesn’t have sway over you like it once did, and that is for the best. Since, now, you truly are a multi-dimensional thinker, relying on a growing repertoire of skills.

Part 1: Emergence, Entropy and the Universe

Emergence is the phenomenon of properties of a system arising (or emerging) due to the interaction of parts in a wider whole. Think of these as abilities unlocked when large-scale network effects come into play. An example of this is consciousness. Today, we understand neurons pretty well, their structure, how they fire electrically and chemically, and how they connect to neighboring neurons. Despite this rich detail on a single neuron, we do not understand how consciousness emerges when 100 Billion of them interact in the structure we call a brain. 

An example more in our wheelhouse are computers. Transistors are miniaturized electric switches used to represent binary electronic switches, the most fundamental of them being AND, OR and NOT gates (see reference #1). NAND and NOR are universal gates because all other gates can be derived from them. From the absolute simplicity of a switch that has two states, "0" and "1," emerged ways to store data, process it algorithmically and distribute it over large-scale networks of the internet. Recent advances have led to surreal applications of Artificial Intelligence and Machine Learning Models that can discern language, drive autonomous vehicles, and help us stay in touch with people across the world. It can even go so far as to model two black holes crashing into each other (Two Black Holes Merge into One Reference #2). From binary logic gates thus emerged human wrought intelligence. 

There are other examples of emergence; 26 letters in the alphabet creating an infinitely extendable communication medium we call language. The four basic compounds called base pairs (abbreviated as A, G, C, and T) interlock in different combinations to form the double helix of DNA. From that aperiodic crystal emerges the complexity of life (base pairs of DNA Reference #3). The four base pairs of the DNA encode 20 amino acids that combine to form inter and intra cellular machinery we broadly call proteins. A cascade of firing neurons leads to thought. There are many more examples of emergence, where deceptive simplicity leads to unfathomable complexity. 

The universe too, in all its complexity, emerged from the interaction of elementary particles constrained by fundamental universal laws. Theory of General Relativity is an example of such a law. One of the most fundamental of such laws is the one concerning entropy, which simplified states that the amount of disorder in a closed system always increases. Entropy is incredibly fundamental to the universe, with Einstein once remarking that 1000 years from now, our species may discover new laws that overwrite current ones but not the second law of thermodynamics. Stated more generally, the second law of thermodynamics states that the universe is inevitably heading towards states of higher disorder i.e. higher entropy.

(In this essay, the terms entropy and disorder are used interchangeably. Higher entropy and higher disorder mean the same thing.) 

Let's take a slight detour to understand entropy, because of how fundamental it is to the universe, and as we will see to life and the mind. Entropy is a statistical representation of disorder. To explain this, let's take 100 standard coins with Heads and Tails and toss all of them at the same time and record their configuration when they fall. The sample of possible states is every possibility between (100H, 0T) to (0H,100T) with the middle case being (50H, 50T). Since each coin can be either an H (head) or a T (tail), i.e 2 possible configurations and there are 100 coins, so the total number of possible configurations (including repeats) is 2^100. 

The above picture which maps out the probability space of 3 coins hopefully gives you an intuition on how quickly the complexity grows.

Let's just take a pause and wrap our head around this number 2^100, which is 1,267,650,600,228,229,401,496,703,205,376. This is a massive massive number. To put it into better context, imagine you have a standard paper that you fold into half, then you continue folding the same paper in half again for a total of 100 times. How thick would the resulting paper stack be? The answer, a mind-bending 13.7 Billion Light years, that is to the edge of the observable universe (see how this is so here reference #4). Fun fact, 42 folds of the paper gets you to the moon. Only 58 more gets you to the edge of the observable universe, which is unreal. A human brain does not easily grasp exponential growth, watch this (provocatively titled) video to build an intuition "The most important video you will ever see. (reference #5)"

Coming back to our experiment, of the total 2^100 possible outcomes, only one branch in the outcome space corresponds to 100H, 0T. Similarly, only one outcome corresponds to 0H, 100T. Hence the probability of such an event randomly occurring is 0.000000000000000000000000000000788 (7.88e-31) which is infinitesimally small. This is the most ordered state, and per the 2nd law, the configuration of the universe of these particles will tend to higher disorder, ultimately converging around the most probable of all outcomes 50H, 50T. You would be very surprised if while doing these random tosses you see an outcome of 100H, 0T or 0H, 100T, but you would hardly bat an eyelid if some configuration around 50H, 50T showed up. Your brain intuitively grasps the improbability of highly ordered configurations. 

Ordered configurations in fact are energy-rich, because building order takes energy, and breaking that order yields energy. In the above example, if you landed with a 100H, 0T configuration, that is the most energy-rich configuration of those particles, and as you extract energy to do other things the configuration will progressively tend towards 50H, 50T, the least energy configuration. So in a simplistic essence, order is correlated to energy. How did order even emerge then if disorder is significantly more probably, and how do you explain stars giving off massive amounts of energy? This is a fair question and one that physics helps answer. 

Without going too much into detail, when the Big Bang happened, fundamental particles exploded from a compressed, incredibly dense core and flew out from which particle clumps started to form. When some of these clumps reached a critical mass, they started exerting gravitational pull and started pulling other particles towards themselves thus transforming dispersed gaseous giants into stars. Stars have a highly ordered core, tightly packed elemental fuel sources that are being fused together due to the intense pressure of gravity. But the further away from the core you go the more disordered the particles are, such that the 2nd law of thermodynamics still holds. The entropy decrease in the core is more than sufficiently balanced by an increase in the entropy towards the outer edge of the star. The sun too, if you observe, has a solar atmosphere at the edge that becomes visible during a total solar eclipse, which is the high entropy part of the star (The Sun's Atmosphere - reference #6). Many years later (~4 Billion years) and a rollercoaster journey, the sun will burn out and settle into being a dead star, entropically equivalent to a 50H, 50T state. 

Part 2: Entropy and Evolution, two sides of the same coin

What explains the emergence of life then? Life emerged from evolution, but why was evolution even allowed to occur? This is where things start transferring from the metaphysical to the personal. Human beings are a byproduct of evolution; we are highly ordered cellular structures and we are blessed with the ability to reverse local entropies to our benefit. The phone you hold, the laptop I am typing this on, a chair, and almost everything we have invented is an ordered structure. Do we break the 2nd law of thermodynamics? No, nothing breaks the 2nd law and as we will see, life favors entropy's thirst for disorder. So much so, that is willing to let order some order emerge so long as the net disorder is always on the rise. 

Evolution is in some sense the twin of entropy, the yin to its yang, the heads to its tail. Entropy allows for evolution to occur, in fact, encourages it. Ordered life forms emerge because the universe wants to reach its entropic potential. In slightly more lay terms, you know when you say that kid has great potential but they did not apply themselves and hence squandered their gifts away? Or that other kid who worked really hard to achieve their full potential, and then did. The universe is kind of like a kid with the potential to do great things, but the universe has to work hard to achieve its full potential. Potential here is analogous to change in entropy, literally how much entropy changed. If the universe stayed in the same configuration that the "Big Bang" wrought about, over vast swathes of time (10^100 years) the energy would eventually decompose but short of star or other mega-structure collisions, entropy doesn't change too much. Entropic potential is achieved, when the energy in existing ordered systems is leveraged to impose order on unordered (random) configurations of particles. Since we know ultimately everything tends to disorder, these newly minted ordered structures too will dissipate into disorder. Additionally, to create a certain amount of order in the universe, you need to expend more order (energy) than you are creating to satisfy the 2nd law. Thus the net entropic change increases, and the universe moves closer to achieving its entropic potential. In a sense, the universe is trying to maximize it's entropic change.

Of all the processes that could enable a universe to achieve its full entropic potential, few work as consistently well as evolution, maybe only beat by evolution's own progeny, an intelligent and conscious species. How does evolution help? Evolution is a process that is enabled by the existence of favorable conditions, the most important ones being - the presence of water in liquid form, a specific temperature range, gravity to prevent the atmosphere from escaping into space which creates a greenhouse effect leading to warmer conditions and thus leads to a cascade of favorable conditions for evolution to start. These conditions are so ubiquitous in our understanding of life that we look for habitable planets using similar metrics (see reference #7 Goldilocks zone). 

The existence of these conditions (a statistical anomaly by themselves) and the presence of carbon and select heavy elements in the earth's core lead to particle interactions, leading to agglomerations with a curious property - their motion cannot be explained by the laws of physics, they seemed to be able to move non-randomly. This ability to non-randomly move and exert control over their own agglomerations is a characteristic feature of life. The earliest agglomerations were single-cell organisms that didn't move much (if at all) but were indirectly responsible for all of life as we know it. These early agglomerations we call cyanobacteria used sunlight to fix the carbon dioxide dissolved in water and released Oxygen (O2) into the atmosphere (reference #8 Cyanobacteria and the Formation of Oxygen). Through a slow process that took billions of years, these ordered agglomerations leveraged the energy in sunlight and through complex metabolic pathways, make for themselves "food." Food is a colloquial term for energy, which a living thing needs to sustain its ordered structure. The need for food to survive is the second core characteristic of living things. 

(Sidenote: If what cyanobacteria do sounds like what plants do, then you are prescient because cyanobacteria form the basis for photosynthesis. In the tree of evolution, photosynthesis was invented only once, every other organism that uses it has cyanobacteria to thank.)

Let's take a moment to see how entropy likes all of this. Even the simplest organism has an ordered core (the body) that is low in entropy but needs to consume energy. This energy is obtained by converting low entropy structures to high entropy, which releases energy, that is consumed as food to maintain its low entropy state (reference #9 Entropy and Biology Photosynthesis). In photosynthesis specifically, cyanobacteria are able to take advantage of the sun raining down energy that is perennially being produced in adherence to the 2nd law. Evolution by virtue of its ability to create ordered life forms, is appeasing entropy's endless thirst by utilizing order but compensating the universe with a much higher increase in disorder. 

This oxygen-rich environment (first oceans and then the atmosphere) was critical to support the next level of complexity in life. The tree of life started bearing the fruit of complex multi-cellular organisms, one of those fruits being the human species. With an increase in cellular complexity, came an increase in energy requirements, which we have established come from ordered configurations. So evolution started imparting species with means to recognize these ordered configurations because it was critical to survival. These newly minted configurations endowed by evolution were responsible for enabling living things to identify energy-rich sources (food), and to chart a path towards these sources. These new structures have names themselves "brains." 

Part 3: The emergence and evolution of brains

The most ordered structures out there to consume were, unsurprisingly, other life forms. Identifying life is not an easy classification to make. Today, we have a good sense of what is edible and what isn't, but when the first organisms were starting out, they needed to be able to differentiate between life and non-life, to classify as edible or not. Let us stop to ponder the complexity of this question, how, without taking for granted hindsight, would a biological brain have been able to differentiate between things that live and those that didn't? Entropy again, I argue, provides the foundation for that answer. 

Entropy is a descent into disorder, and as we covered before, an increase in entropy is a fundamental truth about the universe. (In fact, an accepted theory of time is that time IS the direction of entropy increase.) If everything is tending to disorder, then change to disorder is a constant, whereas life is a structure that maintains its order. In scenes observable to life forms, disorder tends to emerge from the interaction of particular configurations with the environment. The brain against a backdrop of disorder, keeps track of life as the structures that has a few common properties - their structures don't disintegrate into disorder, and the particles seem to be able to engage in self-directed motion. Structures that fit this bill maintain their entropic state, through configurations that persist despite interaction to atmosphere, and they seem to move non-randomly. Dead life, on the other hand, has the entropic configuration of life, but is losing its entropy quickly, decomposing to interaction with the environment, and is also incapable of self-directed motion. 

Thus as an early step, brains evolved an ability to classify. Classification is the ability to distinguish an object from its background and identify the characteristics that demarcate it as such. Or in other words, in the vast web of complexity that comes from particles interacting with each other, classification is the ability to isolate particular configurations, based on common sub-particular configurations that we call properties. 

As an example: We can use the general properties of living things classify a chimpanzee, a fish and a human as living things. But because we are able to distinguish properties, we realize that a human and a chimp are more similar because they live on land, and the sub-configuration of their limbs reveals a common architecture of two hind and two fore-limbs used for moving the body.

Brains, however did not evolve in isolation. Most complex multicellular species evolved brains and in order to preserve themself, they realized that not only do they need to identify entropy to consume as food, but also realized themselves as a source of such energy to animals higher up on the food chain. Evolution selected for brain characteristics that additionally optimized for self-preservation. Now not only did brains need to mobilize bodies to chart a path in three-dimensional space towards energy, but now towards moving targets, food sources actively trying to preserve their own lives, cognizant of an approaching predator. 

This is incredibly complex when you stop to think about it, and huge complexity gains in the brain were needed to be made to support these battles of "eat, be eaten." (This may have been the first time differential calculus was unconsciously applied to close a gap, and integral calculus to anticipate the gap close and increase their chances of survival). This finds resonance in the human world too, incredible technological strides were made during battle. In fact in Guns, Germs and Steel, a seminal book, it was argued that Europe was able to progress technologically rapidly because of nations sharing borders on all sides resulting in constant skirmishes. Battles have evolutionarily sharpened our brains, and we see the resonance of that in our anthropomorphic history. 

However, brains were not the only way to gain an evolutionary advantage in the hunt for food. Size and strength, became important evolutionary advantages. If an organism had the advantage of size (as in the case of dinosaurs, blue whales, sharks) or strength (as in case of lions, sharks, crocodiles) they didn't need to think too hard about getting their food. Their size gave them a vantage point that enabled them to look far, and their strength meant they could take down any food source without too much trouble. Essentially, size and strength put the organism higher up on the food chain, which reduced their survival risks. Hence, lacking in natural enemies, they did not need to breed aplenty to keep their numbers, nor did they need to become ever better-thinking machines. (Reference #10 Brain to Body Mass Ratio)

Humans are at that size where they have several potential enemies that are larger and stronger than them, but not so small that they were in a constant cycle of "be eaten." Human beings were a fairly reproductive species, as up until a few millennia ago, mortality rates of youth was quite high at 46.2% and so was infant mortality at 26.9% (Reference #11 Mortality Rates Of Children Over The Last Two Millennia), this meant the human species was a lot more genetically active than the other species higher up in the food chain. DNA is such that during the life of a species, information can only ever be translated from DNA to RNA to proteins, and never in the opposite direction (this is called the Central Dogma of Biology - Reference #12). This means that changes in DNA, which were the only way for a species to evolve, could only happen via more births. It is not possible for you to change your DNA during life and pass those changes to your offspring.  

So humans were in a sweet spot, where they were sufficiently small to reproduce quickly enough that their genetic diversity was constantly changing in the early days and not so small that they were constantly under threat of extinction from having a natural enemy that marked them for food. In a sense, they were the "upper middle class" of the food chain, due to which they could as a species focus on self-actualizing pursuits like forming societies, growing their own food and inventing religion and nationhood to bind societies. I cannot help but notice how similar this is to Corona Virus, where it was only mildly deadly, but spread easily allowing it to genetically diversify. It too was the "upper middle class" of flu-like viruses.   

Due to human's size disadvantage, they made up for it in massive gains in the capacity of their brain. With nearly 86 Billion neurons, we had more than we needed to just survive. Most of the brain of any organism is devoted to keeping its various life processes running. This, referred to as the subconscious, is what keeps you breathing, your immune system functioning, and your musculature able to control your body (to name a tiny subset of functions your brain performs). Certain functions like the heart beating is (interestingly) brain independent and is carried about by a specific type of muscle that is unique to your heart called myocardium (Reference #13 Cardiac Conduction). Most organism's brain capacity asymptotes at the point that it is able to carry out all these base functions and keep an organism alive and propagate its species. Hence most (if not all) species are highly "present," which means they live in the moment. So if they get hungry: their brain figures out a path to fulfilling that need; they reach sexual maturity: then their brain figures out a path to reproducing; they sense a threat from a predator: then the brain figures out a way to save itself. This is why dogs are so sad to see you leave, and so happy when you come back. They don't have a consistent sense of time elapsed, so when you leave, unless habituated, they have no idea when or if you are coming back. 

Once the base processes were accounted for, the excess capacity of the brain is, for an evolutionarily first time, available for "luxury" purposes. Luxuries that led to superpowers like the ability to sense the passage of time, the twin miracle of speech and language, and perhaps the most powerful of them all - reflection. Our ability to reflect on our thoughts (also called metacognition) is, I would argue, at the center of most of the technical progress we have made as individuals and as a society. It is that last superpower that is our window into entropy, and by extension our theoretical peephole allowing a glimpse into the fabric of the universe through such mental concoctions like quantum physics, string theory and general relativity.  Reflection is also what allows us to observe and deconstruct our own biology. 

Part 4: Brains as a reflection of the universe

Our discussion of entropy touched on how the order of stars came to form from the "Big Bang" while respecting the 2nd law of thermodynamics (entropy always increases). Stars are enormous, mind-bendingly large hearths that not only transmit light and heat in the form of radiation that enables photosynthesis and thus life, but equally importantly, also act as a massive forge to combine the simplest element of Hydrogen into complex elements all the way up to Nickel.  

(The stars are the crucible that combines the basic emergent element of Hydrogen (1p (proton) and 1e (electron)) into more complex compounds, starting with Helium (2p, 2n (neutrons) and 2e) and going all the way to produce Nickel (28p, 28n, 28e). 

Starting from the time of the Greek in the west, we used to think of atoms as the smallest unit, indivisibility conferred to it in its name. We have later come to understand that, atoms are made of protons, neutrons and electrons. We now know even those have more fundamental constituents called quarks. The universe at this layer of abstraction is made of electrons and quarks; quarks constitute protons and neutrons (a proton and neutron can transform into the other). 

(Sidenote: String theory is an even deeper layer of abstraction that posits that the most fundamental particles are not quarks and electrons, but actually infinitesimal strings in extremely high tension that generate all the subatomic particles and forces of the world through string vibrations)

So the universe started off as an agglomeration of electrons and quarks (also photons); quark agglomerations manifest as protons and neutrons; proton, electron, and neutron agglomerations manifest as atoms, atom agglomerations manifest as molecules, and carbon-based molecular agglomerations manifest as the molecular machinery that we call proteins, and proteins manifest life. At every step of the way, particles associate into larger structures, obeying the rules that inform their mutual interaction. But at every stage, what emerges is fantastically different from what came before it. We have almost no hope of predicting an emergent property.  

If you try to visualize a simplified universe, with only 25 atoms (that formed from prior quark, electron agglomerations that we will ignore for simplicity) and 4 types of atoms that agglomerate into molecules. These molecules then combine amongst each other to form proteins, and proteins combine with an energy source in the form of a molecule, and together in our simple 4 atom universe, they are the simplest life forms.

------------------ A simple 25 atom universe ------------------ 

Let's do something to simplify this universe even further. Let us hold constant the same entities, and represent them as one. So all the green molecules, we will represent as one, and draw lines from all the atoms that went into forming it. This way we keep track of individual atom starting points and destinations, however we obscure away the details of the exact path they took. We repeat this at every level and the simplified structure looks like this. 

------------------ A universe with similars grouped  ------------------ 

This looks suspiciously like a neural net, which is a collection of artificial neurons and form the basis of how we program intelligence into machines. 

------------------ Picture of a neural net  ------------------ 

Our neurons have two sub-types of neurons, one is called a pyramid neuron (because of its shape) and another is called an interneuron (or an association neuron). Pyramid neurons look remarkably like the path of one of these particles in the forward direction. Purkinje cells are another type, but they are in essence a pyramid neuron with significantly higher nerve density. 

------------------ Pyramidal neuron - courtesy Nature.com  ------------------ 

Interneurons, or association neurons, are what discover similarities and "associate" them all. When we held all similar molecules constant, we were able to do that because our association neurons noticed the similarity and allowed us to hold that feature of the system invariant, i.e. hold a feature of a system constant to simplify the mental representation of the complexity. We would be overwhelmed by the complexity of the world (emerging from entropy) if we lacked the ability to hold something constant. Association neurons are also vital to improve computational efficiency (see Appendix for an explanation and an analogy).

------------------ Interneuron - courtesy School of Biomedical Sciences  ------------------ 

Is it not remarkable how analogous to one another the universe and the brain are? The brain is quite literally a reflection of the universe, and maybe that is how we perceive the universe indirectly, first through the senses and then by recreating it in our brain. While most animals do the same, what makes us different is that a part of our brain is sectioned off to store a core pattern, our pattern of self. I talk more about it in another essay but the core idea is that our identity comes from a set of memories and influences that shaped the brain. Memories are but a reflection of a slice of the universe that we call our environment. In recognition of the importance of these memories it stores them separately and ties them together into a super pattern. Since our brain loves telling stories (the story of the universe for example) these memories and influences when linked together lead to the emergence of a central story. This is the story that defines our sense of "self." Since any memory is of the real world, and the only thing that is capable of interacting with the real world is not our brain but our body, our bodily perception are a critical part of our "self."  Our sense of self is the part of our brain held invariant, that is able to experience, witness, and influence the other parts that change and are free to change.

Part 5: The brain, language and entropy

This brings us to language, one of the most beautiful inventions of the human mind. It is impossible to know the origins of language because language originated as far as we can tell in the human mind alone. Our records of what we can loosely call language started off as cave paintings, followed by hieroglyphics and many intermediate forms before formal writing came along. Some writing survived the ravages of time and helped us piece together a history of language's emergence. The word emergence is important here, and it is a reinforcement of a pattern we talked at the very start of this essay, a property of a system arising from the interaction of parts in a wider whole. Here the interactants are human brains, and the emergent property of the network of brains is language. 

While we can never be sure of language's exact story of emergence because of it's inverse nature (Reference #14 Inverse Problem) we can piece together a story by studying its effects. We know that there is overlap in the areas of the brain responsible for language and action (motor) area (Reference #15 Brain Mechanisms Linking Language and Action). This helps identify a causal relationship between the two, with hand gestures like grasping having come first followed by a linguistic representation of that gesture. Are animated gestures when speaking are a resonance of the overlap. 

How do humans converge towards a particular gesture or a word that denotes it given the infinite variations that are possible? This is where network effects come in. Since language of one i.e. a language that only you understand is not very useful in communication, there needs to be a wider consensus. This is naturally achieved in societies through the wisdom of crowds, which comes close to optimal solutions. Let me give you an example to expand on this. In grad school our behavioral economics professor showed our class of ~60 students a picture of a big cow, and asked us to guess it's weight. We had no additional information beyond that picture, and we were all to write our answers on a piece of paper and pass it along to the TA who collected and tallied it. What came out was memorable enough that I remember it to this day many years later. When tallied the answers had a wide range, with comically low and absurdly high guesses, but curiously, when averaged out, the final answer was remarkably close to the actual weight of the cow, within ~2% of the actual. This is not a one-off result but rather repeatable, where every class before us that the professor had conducted this experiment on were within ~5% of the actual weight. This is the underlying principle behind democracies; if the people are informed and free to independently choose they are likely to arrive at the most optimal candidate.

Even in lab experiments where people were asked to choose gestures for a particular set of words without speaking, first individually and then collaboratively, in a matter of 10 group iterations, the gestures for the words came out to be consistent both within and across subjects (Reference #16 Evolving artificial sign languages in the lab). You can think of language as having emerged from many such iterations where the initial gestures and sounds to denote a particular action was random but through every brain it inhabited, language was refined and grew closer to the most ordered representation of sounds stringed together to represent something in the real world. The birth and death of languages (put picture of language tree here), helped more refined and finer grain order to emerge. 

------------------ A Language Tree - credit SSSScomic  ------------------ 

This is not to say that language is all structured, in fact, most of our word choices are completely arbitrary. Why the word banana for the fruit vs any other combination of sounds? It is likely somebody must have come up with it, the word slowly became popular with others and after a critical mass of people started using it, the group collectively agreed. In today's time, dictionary companies perform this function by accepting new words into the language every year. However, what collectively came out was a common syntactical structure to language lacking which we wouldn't be able to understand it uniformly. Our understanding is also helped along by fixed words like pronouns (he, she, they, I, you, etc) and determiners (the, a, an, etc). These parts of language don't change every year, they are largely invariant. You can maybe start seeing the similarity between evolution and language, where language is the emergence of information from symbols. A natural question to ask then is what is information?

Part 5.1: Information, association and entropy

While it may sound tautological, information is the ordered configuration of symbols that refers to entropy in the universe, or, more specifically, the aspects of the universe that reduces "referential entropy" or "interaction entropy." These are new terms and let me explain what I mean. 

Referential entropy is the name given to associations in the entropic space like we discussed previously. Association of quarks and electrons are called atoms, associations of atoms are called molecules, some associations of carbon-based molecules are called life and so also is a chair an association of particles where we can comfortably sit on and lean our back on. 

Interaction entropy is the name given to change, the changes that caused a "referential entropy" to move in spacetime or to change it's agglomeration structure. This is difficult to easily explain (probably because I too am not very clear) but think of it as the name given to the links that lead to particle agglomerations to change their location, or their configuration. Verbs for example represent interaction entropy, where move, push, walk, or breathe represent changes to locations of agglomerations. Words like force, gravity, heat are a quantitative measure of the interaction entropy because they are able to explain how particles will interact without waiting for the interaction to play out. When something is decomposed, built on or torn down (at the atomic, and higher levels of agglomeration) is a form of interaction entropy that leads to new referential entropy.

------------------ Referential and Interaction Entropy  ------------------ 

I gave this somewhat involved explanation to arrive at the point that information is the notation given to the order in the disorder (or entropy) of the universe. The law of gravity is incredible because it has the capacity to take almost any slice of the universe with objects of referential entropy (planets, stars, space shuttles, and every particle agglomeration past a certain size) and predict its entropic interaction, i.e the interaction entropy. In fact we have a mathematical law that reveals a fundamental force that is able to hold its own against entropy. 

Similarly relativity is magnificent because it was able to identify a bound to the universe, that the speed of any object with referential entropy cannot exceed the speed of light, which is an interaction entropy. It was essentially able to prune the tree of the universe by invalidating a set of interaction entropies, the entropy achieved of particles moving at speeds greater than the speed of light. By reducing the entropic space of possibilities, Einstein was able to follow other branches and realize the curving of space. I do not mean to trivialize the achievements of Einstein in discovering this. To be able to model the universe and it's entropy on such a titanic scale and derive the central order from all of it is unparalleled in its creativity and cognition, something only a human brain could do since it is the reflection of the universe. 

It makes sense now why language is so difficult for computers to understand because when we speak a sentence, every word plays a part in progressively reducing the entropic space of the universe in focus. We also call this attention. Language is also inherently recursive, we can write a sentence and then in the middle of the sentence choose to add detail about a specific part, kind of like I did with this fragment right now. The recursivity of language is an important feature, because it allows us to use language to talk about language. It is an excellent reflection of consciousness, central to which is our ability as a thinking being to look inward and reflect on our own thought. No wonder humans were the ones to come up with language, our brain reflected itself out into the real world in the form of language. 

Part 5.2: An aggregation of abstract entropy

This is an interesting avenue lets dig a little deeper by considering a trivial example. Take the example of the word "chair" and its conventional definition of being a place we can park our buttocks on and rest our backs. We associate it with comfort because we transferred the work our body was doing to keep itself upright into the external world, thus saving energy expenditure that kept our muscles contracted. We recognize a wide spectrum of chairs, from human-engineered ones to flat rocks in specific formations that afford us the comfort of resting our muscles and leaning our back. This is a different form of referential entropy (or particular agglomeration), because there is no inherent uniformity. Up until now when we considered objects as having referential entropy, we were looking at increasing levels of associations, where atoms associated to form molecules, molecules to proteins, and proteins to life. The similarity is much higher than in what we refer to as a chair. A chair could be made of wood, steel, rocks, branches and so many other things and it still fits our loose definition of a chair. How then is our brain associating such diversity into a single definition? 

This is where the body comes in. Our brain inherited the capacity to form associations from the universe but then we made it our own. We are not constrained to form associations only of the type that the universe most commonly experienced. We experience the universe through our body, and the experiences received through the sensory inputs of the body are a rich source of potential associations. When Descartes formulated his theory of the mind, he proposed a dualist theory - one where the brain and body are separate, with the body being a sort of homunculus, dumb and stupid on its own but elevated when animated by the mind. (Dualism by Stanford Encyclopedia of Philosophy - Reference #17) The mind was the center of consciousness, the body a necessary appendage to host it. From this rather bacterial definition of life, we have philosophically evolved to a different model where the brain and the body are not separate but a single entity, neither meaningful without the other. This theory called materialism dispenses of the special treatment given to the brain and instead argues that there is no mind, only a body. Genes express and neurons fire in the body, and through that we receive and inhabit the universe. 

Our brain housed safely in our skull, is for all intents and purposes blind to the universe. It only receives it through the senses, then reconstructs it into a simulation and inhabits it. Our body, on the other hand, actually inhabits the world and thus plays a critical role in shaping language. The wide diversity in the agglomeration of particles being called a chair is because the body sees its importance in its interaction with it. So the referential entropy of the body, has an interaction entropy of sitting on a wide variety of objects that are entropically quite different but have perceived attributes of importance. Here the attributes are of comfort to the muscles in the body, specifically your leg and back muscles. 

Perceived attributes are quite important too, they have no entropic significance except that they are a set of abstract attributes that are important to the object that is interacting with it. The beauty is that if the object was lifeless, no perceived attribute would exist and what you see is what you get. However, since life is self mobilizing, and endowed with the ability to act independently (through the consumption of energy) it gets to find its own meaning, or stated more generally, attribute whatever importance it wants to. Since we are still, ultimately, pattern recognition engines, we don't attribute one-off qualities (or qualia) but rather, we converge on attributes that hold common importance. This is an extension of the consensus and the wisdom of crowds' idea, where the attributes that are important evolve from the survival of fittest playing on attributes. These most-fit attributes then associate together into objects of abstract referential entropy. This type of referential entropy is different from what we previously discussed because particles are not actually agglomerating, rather they are agglomerating into mental structures in the mind, vetted by social and cultural standards. 

This idea of an abstract attribute that is only important to us humans also doesn't stop there, as would not surprise you it also agglomerates into higher-level associations. This is different from referential entropy because no actual particles are involved, rather it is a coagulation of attributes. For eg: the word home, is a non-exhaustive combination of attributes of security, comfort, shelter, privacy and safety (most of which are abstract attributes themselves).  

Another category of words are those of physiological needs and emotions, which sometimes go hand in hand. The feeling of hunger, thirst, doubt, sadness, anger, frustration and jealousy (to name a few) are very important to us and come from the interplay of the body and the mind. Physiological needs like thirst and hunger are a body communicating its needs, while words like comfort and safety are a body's response to the attributes of the part of the universe their body inhabits (what we also call environment). 

Emotions are very interesting in this context, because they are the seat of so many things that makes humans creative, thoughtful and reflecting beings. We discussed in an earlier part that our ability to reflect is one of our most powerful abilities and separates us from most of the living kingdom. We established that we perceive the world through our senses and then recreate the senses into a representation of the universe, or what we call reality, and the brain perceives us inhabiting it. Thus we rarely receive the universe as is, we color it with our biases borne from our memories and other facets of our "selves" that were impressed upon us by the very universe we live in. So we simulate the universe as a personal reality and inhabit it. All of us thus inhabit our own slightly different realities, which is why our languages are slightly different but close enough that we can communicate. We will use the word "truth" to refer to the uncolored universe, which I don't even know is possible to mentally receive. We need a control function then to keep our reality simulation in check, otherwise what is to stop us from creating fantastical universes, completely different from anything remotely real and inhabiting it. We associate this with mental disease, a child who thinks they have superpowers and can fly, and attempts to jump off the roof. This clearly is very dangerous and needs evolutionary protection. 

I propose that emotions evolved as this control function on universal simulation. Sadness is when our simulation of a universe fails the reality test, and happiness is when it passes the test. Thus, sadness is the simulation of a universe with referential and interaction entropy that is rejected by reality, while happiness in a sense is one that is accepted by reality. Trust is the interaction entropy between two "selves" that has over time proved itself to fit each other's simulation, while distrust is when reality has taught you of the failure of your prediction. Jealousy is when the autobiographical simulation of your universe is superseded by the reality of somebody else's. Humour is a unique interpretation of the laws that lead to a simulation of a universe that is for all intents and purposes possible but either knowingly impossible, or only remotely possible, or plays up an abstract attribute. Many times humor involves a shared, almost telepathically communicated new parallel universal configuration with changed parameters that enable the impossible, like talking animals or the personification of inanimate entities. However, humor hits hardest not when it is completely nonsensical but sensible per the laws of the parallel universe that was mutually agreed to in the shared consciousness of the humans participating in the humorous experience.

Part 6: Free will and our ability to shape reality

The universe plays itself out probabilistically, where no particular pathway in the entropic space of possibilities is favored, but is a cascade of probabilities. For the current world to have the form and structure we see, every past event had to happen in the exact order that it did, and every living thing in it had to behave exactly as it did. The slightest variation has the potential to create a substantially different universe that we may not recognize. This steps into the arenas of philosophers who use the inevitability as an argument against free will. If everything that happened had to happen for the current reality to manifest, then are we but biological machines playing our fixed roles in the entropic canvas of the universe? Doesn't this presuppose the existence of a god or a higher entity that carved out our roles, or is it just that this entity set in motion the first few cascades, and all of reality is a follow-through of that cascade? 

I will argue that we do have free will, but it is a choice we have to intentionally exercise. I will also argue that free will that can shape reality is a function of aggregating (or agglomerating) probabilities. Simple animals without the ability to execute complex coordination do not have the weight to pull off large-scale changes. This is not to say that they can have no impact. If a pack of wolves hunted every rabbit in their area and if hypothetically the rabits only existed in that one area, then given enough time the wolves would have whittled down the population of rabits to a sparsity that they are unable to find each other to reproduce. In effect, this pack of wolves drove a species to extinction. This is not dissimilar to what human poachers do, driving the populations of elephants to near extinction. 

Human beings when they are born spend almost 18 years in development before they are considered capable enough to become independent adults. During this time of development, the brain is shaped by the immediate environment. We are born with near blank slate brains but equipped with incredible neuroplasticity to drink in our surroundings and start modeling it in the brain. Our parents, our upbringing, our friends, and later, the society, all play a significant role in how we turn out. Each one of them fills us up with their own biases, and their idiosyncratic way of modeling their universe. Our brain has incredible machinery to distinguish features of the universe. We talked about our ability to classify in an earlier section, and this ability is something all of us are born with. However, we can only work with what we have been fed. So our universal simulation is an aggregate of the universal simulation of the people who influence us (with different weights). If we continue in this state, then we exit the interstate of universal possibility and step into a cul de sac of our a social simulation. We slowly lose our ability to exercise free will. 

However there is a way out, and that is to feed the brain with experience, or what I pedantically call new slices of the universe that the brain could not simulate by itself. We use various language analogies to describe this, including to "broaden one's horizons" and "expand one's worldview." This is why a child who grew up in a rural village in India, who only went to school in the village and was taught by the same teachers that went to the school a decade ago, will rarely make life choices outside the norm of that village. However, a child who constantly experiences new cultures, is surrounded by diversity in a bustling metro, spends time in a culture diametrically opposed to his, or is bilingual (since we made the connection between language and universe simulation) is able to calculate probabilities on a much wider scale. The neurons calculate probabilities in the universe that the brain as a whole is able to simulate. So a child in the rural village can imagine the possibility of starting a company and becoming rich, but the neurons calculate that probability to be zero since it has never encountered that happening. Whereas the simple act of doing can expand your horizons. 

The experience of doing something that your brain firmly decided was not possible for you, but due to circumstances or "will power" you powered through and emerge unscathed on the other side is powerful. The brain learns to place less emphasis on its own simulations and realizes that a lot more is possible. As an example say you did not think you were capable of learning to program, but due to being laid off you were forced to pick it up and a few months later you end up with a programming job that you are much happier at. Now the brain realizes that a lot more is possible than you thought and the next time you have a very low probabilistic thought, like starting your company, your brain doesn't automatically shut it down as impossible.

The brain is specific about the types of experiences that can change it, almost always an active experience is required, rarely if ever does passive experience have the same impact. Watching a documentary about somebody building a robot vs you actually building the robot has a huge variance in the impact to the brain. Watching a series about a group of friends in New York City, is very different than actually living in New York City and having friends, no matter how many times you watch it. Knowledge is another way that the brain expands its space of probability. A physicist who understand how the big bang happened, and how quarks and electrons that originated there interact amongst themselves to form planets and solar systems and life, is able to model much larger scales of the universe. A biologist who is witness to the incredibly subtle symphony of molecules and compounds that lead to life is able to model the universe in incredible depth. Having knowledge thus increases the simulated space of your universe in both breadth and depth, and thus increases the scope of possibilities that you assign. You now get to exercise free will, because no longer are you bound by the probability space that you were endowed through your environment. You get to exercise the characteristic feature of life, non-random motion not just in physical space but in entropic space. Which path you chose is a function of your emotions which place a personal value on different paths, and you choose the one that is most meaningful to you.

Now that we understand how individuals gain free will, how is reality affected by individuals exercising their free will? Here again, the concept of aggregation and associations play a central role. While a single individual exercising their free will does not affect reality. However, when a path in entropic space is seen as valuable by others and people gather to support this individual or organize themselves to act as allies or as employees, then the free will of a single ambitious and audacious individual can change the course of the reality. In fact that is probably the only thing that can lead to changes in human timescales. Our objective reality is like a rope made of fibers of all of our individual realities such that through wisdom of crowds, the biases are stripped and only the real parts persist. A small set of people usually act as stewards of reality, and choose which direction humanity as a whole heads towards. We have had luminaries like Socrates, Galileo, Kepler, Newton, Maxwell, Fourier, Euler, Einstein, Schrodinger, Nietszche, Krishnamoorthy, Bohm, and in modern times Steve Jobs, Elon Musk, Bill Gates and Jeff Bezos. We have also had people like Hitler and Stalin do the same.  

We need more people to step up to the role of stewards of humanity. However for that to happen more people need to think outside selfish scales and start imagining on the scale of humanity. For this to happen, you need people who seek diversity, and knowledge and realize that they are a manifestation of a universal consciousness and that their duty lies towards all of humanity and all of nature. 

---- Thank you for reading ----

Appendix:

Association neuron analogy: 

Another reason association neurons emerged is to support more complex functions in animals, like movement as an example. To do something as simple as putting a leg in front of another to take a step involves effecting many neurons that allow fine-grain control over the individual muscles to maintain balance and coordination. If the brain had to individually compute contractions of every single muscle fibre, it would take massive computing capacity to do the simplest things. Association neurons help increase computational efficiencies by connecting up groups of nearby muscle fibers into an association neuron. Now to move a leg, you activate the respective association neuron and it will automatically activate the muscle fiber groups that it connects to. (see appendix for a good analogy)

Let's take the example of a marionette to drive home this point. If you start off with a simple marionette with 3 strings to control it, this is fairly doable and within the limit of the digits on your hand. As you keep increasing the number of strings, the complexity of managing the marionette increases. With training, you can manage 10, with one string per digit using both your hands. Now imagine the marionette evolved in complexity, and grew to 100 strings. Due to the increase in the number of strings, the marionette acquires more exact and precise movements, but there is no way you can handle 100 strings. You don't have the required number of digits. The only way to still work the marionette is to group strings and reduce the number to a max of 10. Grouping the strings that work the hand of the marionette will lead to much smoother animation of the marionette, than say grouping some of the hand strings, some of the leg and some of the head together. However, these groups can then form a supergroup that is involved in a coordinated movement like walking the marionette, an act that involves the hands, the legs, and the neck. 

Reference: 

1. Logic Gates: https://www.circuitbasics.com/what-is-digital-logic/
2. Two Black Holes Merge into One: https://www.youtube.com/watch?v=I_88S8DWbcU
   
3. Base pairs of DNA animation: https://www.genome.gov/genetics-glossary/Base-Pair
   
4. 100 paper folds to the edge of the universe: https://www.freemars.org/jeff/2exp100/question.htm
5. The most important video you will ever see - https://www.youtube.com/watch?v=F-QA2rkpBSY
6. The Sun's atmosphere: https://scied.ucar.edu/learning-zone/sun-space-weather/solar-atmosphere 
7. Goldilocks zone - https://exoplanets.nasa.gov/resources/323/goldilocks-zone/
8. Cyanobacteria and the Formation of Oxygen - http://butane.chem.uiuc.edu/pshapley/environmental/l30/1.html
9. Entropy and Biology Photosynthesis - https://www.ecologycenter.us/population-dynamics-2/entropy-and-biology-photosynthesis.html
10. Brain-Body Mass ratio - https://en.wikipedia.org/wiki/Brain-to-body_mass_ratio#/media/File:Brain-body_mass_ratio_for_some_animals_diagram.svg
11. Mortality Rates Of Children Over The Last Two Millennia - https://ourworldindata.org/child-mortality-in-the-past 
    
12. The Central Dogma of Biology: https://www.yourgenome.org/facts/what-is-the-central-dogma
13. Cardiac Conduction and Myocardium - https://www.cliffsnotes.com/study-guides/anatomy-and-physiology/the-cardiovascular-system/cardiac-conductionzoo
14. Inverse Problem - https://en.wikipedia.org/wiki/Inverse_problem
15. Brain Mechanisms Linking Language and Action - https://www.researchgate.net/publication/7784335_Brain_Mechanisms_Linking_Language_and_Action
    
16. Evolving artificial sign languages in the lab: from improvised gesture to systematic sign - https://psyarxiv.com/be7qy/
    
17. Dualism explained by The Stanford Encyclopedia of Philosophy: https://plato.stanford.edu/entries/dualism/

At the moment of writing this, I am sitting at an airport looking out onto the runway, at planes landing and flying, at the harmony of airline crew members making sure that the aircraft are tended to and supporting a timely departure (I sound like an in-flight announcement don't I). I cannot help but marvel at how far human ingenuity has brought us. Not very long ago (a blink in time), we were hunter-gatherers living in forests and running around trying to gather food and keep ourselves safe. Long before that (a few billion years ago), we were single-cell organisms living in an ethereal soup that the early earth was. In those days of single-cell organisms, the first multicellular organism emerged from the symbiotic relationship between a cell that consumed other cells as nourishment and a smaller, simpler cell that converted free oxygen into energy. In an act of defiance to their design, the larger cell, instead of consuming the other cell (like the norm), absorbed it. So now, the smaller one could continue to produce energy by fixing oxygen, while the larger one formed a protective membrane around it. This was the birth of the multicellular organism, and that cell that fixes oxygen is found even today in every cell you possess. It is called mitochondria, and it is infinitely more complicated than that first multicellular organism. Still, it does the same thing; it generates the energy that powers all the other cells in the body. 

Our brain develops using similar pathways, and the cells that constitute it have a very specialized function - to "think" and to "do." The brain evolved to become the control center that coordinates "living" and "being." Other animals have brains too, but I have often wondered what makes us human beings so different. Let me try to think through that in this essay. 

While other animals have brains, what makes us different is the presence of a pre-frontal cortex (PFC). In many ways, the PFC is a reverse evolution engine and equips us, humans, with the embedded learnings gleaned from billions of years of evolution. Think about it, evolution is a relentless struggle against chaos, the most definite way of making order in absolute disorder and reinforcing patterns that survive time. Realizing that my brain is akin to a machine learning model, trained on billions of years of the universe, and the control function being survival, was pretty profound. Apart from pattern recognition, we are also capable of storing patterns in our memory (in a way that is the only thing we store). Thinking of the brain as a pattern recognition engine and our memory as pattern storage opens us to some interesting ideas. 

When we are born as babies, the underlying architecture of the brain and the "evolution engine" is present within all of us. However, we are not born knowing things; we learn them (very, very quickly, I might add). So in a sense, the first few years of your life are the priming period of your brain. Your environment and your upbringing lead to your sense of self. Our sense of self, is a stored pattern. Our caregivers and our environment shape our perception of ourselves. As we grow, we add and subtract from that pattern. It is what ties your memories together, and how you are the "hero" (pattern) in your story. Your sense of self is a core pattern, and not changed easily or without effort. If it changed too easily, then your brain cannot tie memories together. This is why people's sense of self only changes after a struggle. If it was too easy, then it isn't a core pattern. In this model, "ego" is a necessary side-effect, a sense of self where the pattern is based on extrinsic factors. As your brain is priming itself, it needs external indicators to formulate its own control function. The desire for money, fame, power, and sex is your ego exerting itself, a pattern that your brain latched on to at an early age and integrated it into the pattern of your "self." As you grow, you have to work on your ego - just because your brain latched on to societally desirable patterns does not make them right. Every human being should contest with their ego and shed those patterns that your older, wiser self knows is weighing you down. 

Music is a pattern in sound, and hence so stimulating to our brain. We may not be able to create it, but a beautiful pattern does not go unnoticed. This inherent pattern in music is also why there is musical theory and why when an untrained pianist plays, it sounds noisy, but when a trained one does, it is beautiful. The trained pianist knows to generate a pattern through music, not noise. 

A story is a pattern communicated through words, telling the start, the middle, and an end. The brain's episodic memory is concerned with connecting the cacophony of your sensory inputs into a cohesive story. We love a hero's story because our brain constructs this story for ourselves. We are the hero of our own story, where our sense of self is the main character, and almost everybody else, a supporting cast member. A good movie is an episode constructed by a director, where your brain doesn't need to do the heavy lifting of tying sensory inputs together. Being a good storyteller is hence a gift; you are able to turn words into patterns, and since our brain loves patterns - the coherence of the story can move us, spur us to action, bring down power structures and mobilize masses of people (Mahatma Gandhi, MLK, Hitler).

Knowledge is a pattern of patterns, which is why knowledge acquired through memorization doesn't stick around for too long. Real knowledge and learning come from attaching a learned pattern to an existing pattern in your brain. Unconnected patterns are lost to time. This is also the underlying theory of chunking, where information is stored in your short term memory for processing and vanishes pretty soon; unless you see a pattern in the information and you associate it to that pattern. When you do associate, the information goes into your long term store where it persists. The more patterns you have in your arsenal, the more incoming information you are successful in transitioning to your long term store. 

Science is finding new patterns, and the scientific method is a time-tested system to ensure the veracity of new patterns. This is why I find science so beautiful; you are finding new patterns in the universe, thus reducing the information entropy. An engineer is concerned with taking these patterns and turning them into useful contraptions. For example, Bernoulli figured out a pattern in the way fluids behave, where depending on the fluid's velocity, it exerts different pressure on the material it is flowing through. Modern airplanes are engineered around that pattern, where the shape of the wing forces air (which is a fluid) to have different velocities above and below the wing. The different velocities generate a difference in pressures, which generates a lift (that is why the wing is shaped in a bulbous way).  

Creativity is finding an imaginative way to connect unrelated patterns or see patterns where others don't. Einstein elegantly connecting the pattern of space and the pattern of time through relativity was a feat of scientific and creative daring. 

Curiosity is a deep appreciation for patterns and a propensity to look for patterns everywhere. 

Awe, is when you receive a pattern so profoundly beautiful that you are transfixed. 

Love is a resonance of patterns. 

What a privilege it is to be alive. To be endowed with a neural engine that is a condensation of billions of years of evolution, and we get total and complete control on what to apply it on. Evolution is a beautiful recursion, it is through evolution that the universe is attempting to understand itself. We are gods, and our kryptonite is time. 

Shine on, you crazy patterns. 

Associations: 

• Society of Minds by Marvin Minsky - https://www.amazon.com/dp/0671657135/
• Molecular Biology of a Cell - https://www.amazon.com/Molecular-Biology-Cell-Bruce-Alberts/dp/0815345240
• Bernouli Theorem Explanation - http://hyperphysics.phy-astr.gsu.edu/hbase/pber.html

A friend asked me what I think about the interplay between governments and big-tech and whether governments should start regulating industries more. It was an interesting discussion, and I got the idea to jot down some thoughts and associations that I made. Most of the associations were happening in the economics of Adam West, a general understanding of Marx's theory and how it played out in Russia, and a general underlying optimism that the pursuit of knowledge and act of creation are our ultimate purposes in life. 

Let's start with Capitalism; Capitalism is the form of market economics that incentivizes innovation through wealth creation and private ownership. Communism is the form where communities are the only owners, there is a central government that represents the interests of the communities and they own everything. Communism and Capitalism were the two most subscribed branches of governance and they clashed in a near nuclear attack during the Cold War. Capitalism won out, not just because of the US's higher defense budget but also, I believe because Capitalism was better able to align incentives. 

Communism sounds awesome in theory, no poverty and everybody is able to live a certain quality of life irrespective of their economic contributions. Jobs that existed were in service of the governments, produce was traded to the government at a price set by the government, and the government took care of the distribution. Individuals had no bargaining power, who else were they going to turn and sell their products to if not the government. The government went so far as to make trading with anybody but themselves an illegal, punishable offense. Let's extend that farming example, say you were a farmer, a brilliant and observant one. You notice that the process of plowing the field is inefficient, and you come up with a design for a tractor, a mechanized automaton that would save you 50% labor costs. However, the idea is one thing, execution a whole other thing. Would you as a farmer be incentivized to go out of your lane, invest in this (and seek investment), build and iteratively test it and finally open it up so that others can benefit from it? 

Well, not really right - even if you did all that and managed to build yourself a tractor (a big if), you most benefit from it if you keep the innovation to yourself. If you share it, the government will realize that farmers can now produce more at the same cost and so they will uniformly reduce prices. All your savings vanish and you're back to where you started, the government still takes 90% of your produce and gives you the same total amount. Another related problem that shows up is a fixed mindset. As an example, imagine you came up with a groundbreaking innovation that could generate power for much cheaper but requires an upfront investment to play the idea out. In a communist empire, the government controls energy production. The people who are in positions of power in these institutions are incentivized to maintain the status-quo because the alternative could grow big enough to strip them of their authority. Once you've tasted power, it's really hard to give up. 

Let's play the above situations out with Capitalism, which, as we remember, is predicated on private ownership and subsequent wealth creation. If the farmer came up with the design of the tractor, he/she has all the incentives in the world to go out and raise money from willing investors who see the potential in the idea. This allows the inventor to go build-test-iterate till a fully-functioning and well-researched product is out in the market. The tractor achieves product-market fit, and soon every farmer in the US wants it; soon, word spreads and farmers in India and other countries want one too. In the process, the inventor is rewarded with wealth, status, and an ability to use their wealth to incentivize further innovation. It is more resilient too. In the case of a new cheaper power generation product, even if the government is not interested in funding your pursuit, that tractor owner with overflowing wealth sees the potential in it and decides to take a risk and back you. You now have the runway to build-test-iterate loop and, in the process, come up with a product that cuts the consumer's energy cost by 80%. The market loves your idea because you are saving them money, and as you hit critical mass, the government sees the benefits and contacts you to standardize your product more generally. 

Big-tech is a massive, massive beneficiary of Capitalism. Companies like Amazon were so innovative that they cornered massive markets and in the process, generated enormous wealth, which they redistributed to more innovative profit-seeking endeavors. This is a close cousin of the idea of a perpetual motion machine of wealth creation. There are few industries worth disrupting that Amazon is not in (retail, space, films, healthcare, internet infrastructure, gaming, to name just a few successful ones). I both admire and fear their scale. Amazon is also not the only one in this hallowed category, some of the others worth mentioning being - Microsoft, Facebook, Apple and Google in the US, China has Alibaba, Baidu, Tencent and (newly) Bytedance. Left unchecked, these institutions may grow to become more powerful than a country's government, reaching a point where the power imbalance means that the US needs Amazon more than Amazon needs the US government on their side.

This is the part where benevolence comes in. An important point to remember is that tech induced wealth creation was so tremendously rapid that Amazon has generated $200B+ of personal wealth for Jeff Bezos in just over 25 years (started in 1994). These are all single lifetime wealths; most if not all of tech is first-generation wealth creation. The people who have stood these companies up and weathered the risk, financial burden, and market economics to become successful are uncommonly visionary (we can never have enough visionary people but that's just me). They also seem to be rational humanists who intend to use their acquired power to advance the human species. Pretty fortunate that these people are benevolent and out to do good right? Well, not exactly, and let me explain my rationale. 

People who generate outsize successes, the kind that vaults them to the most successful or the richest, are a self-selecting set. The market acts as a forcing function, selecting those individuals who solve a problem on a wide enough scale and do it for impact rather than monetary gains. The bad apples, the Berni Madoffs, the Elizabeth Holmes and the Martin Shkreli's of the world, can sprint ahead but reality usually catches up. It's like the "great filter theory" of evolution (technically abiogenesis), used to explain Fermi's Paradox but for wealth. I do not know of the specific filter that catches the no-gooders and the insufficient visionaries, but it makes associative sense. It's almost like these people should be incentivized to run with their creativity and bring an opinionated world view into existence (see my post on The genius in world building). This is important for humanity's continued existence, or we may be wiped out by a great filter. So in a sense, the existence of these mega-companies with world-changing ambition is our attempt to create knowledge and stave off the great filter. 

Ok so, the market chooses benevolent people to succeed, sounds like a win-win then? Well, no, not really and for a few reasons, but there are two main ones we will be focusing on. I'll write about that in the second and final part which I'll finish up soon. 

Association points: 

• The Wealth of Nations, Adam Smith 
• Marxism and subsequent Communism
• How Innovation Works, by Matt Ridley
• Capitalism and Socialism: Crash Course World History
  
• Fermi's Paradox - If the universe is infinite where is everybody else? 

Evolution is incredibly elegant in the way it reversed chaos and led to the ordered structured of atoms that we call life. We human beings are the pinnacle of that process, having been given a forebrain that is able to consciously exert its influence, as opposed to the subconscious way most (if not all) other organisms live. The conscious brain (CB) is truly beautiful; it condensed the knowledge gleaned from billions of years of evolution and turned it into a computation engine that almost every human being possesses in various capacities. Evolution created patterns in atoms, and your conscious brain is a pattern recognition engine. 

The subconscious brain (SB) is the ancient brain, the part imprinted by evolution and responsible for keeping you alive and functioning as a member of your species. Things like breathing, digesting, sleeping, swimming (interestingly, we are all born knowing how to swim), and mating, to name a few. You have these subroutines programmed into you before being born; they aren't learned. The process of digesting food, as an example, is amazingly complex. The muscles in your stomach contract and release in a wavelike harmony to push food along the intestines; you generate chemicals to break down this food, convert it into energy (ATP) and excrete the waste. If these are the number of things that need to be coordinated for a single subroutine (digestion), you can start to appreciate how vast the computational capacity of the subconscious brain is. It is astonishingly more powerful than your conscious brain.  

However, it is the interplay between the conscious and unconscious that differentiates human beings (and a few other species) from the vast majority of organisms that exist. Because you will realize, that the CB is able to actively add to the subconscious subroutines. Take as an example walking; you aren't born knowing how to walk, you learn it as a baby by falling many many times before the CB finally gets it. Once it gets it, it sends the learned subroutine to the SB, where it is near permanently stored. The number of muscles engaged when you walk is staggering. Imagine you consciously having to select which muscles to engage when you put a step forward while keeping track of your center of gravity, incoming traffic, tripping hazards, and maintaining a walking speed. 

This is the interesting interplay between the conscious and subconscious brain. Your CB can actively update the SB, and because you can choose what the CB focuses on, you can bring to weigh the incredible computational capacity of the SB on complex routines (like learning how to solve a Rubik's cube or play chess). However, your SB is not a thinker; it is a doer. What that means is when your SB is engaged, your sense of time falls away because the SB is just executing subroutines. 

Think of the last time you drove (a learned subroutine in the SB) with a friend who you were excitedly talking to. An hour later, do you remember many details about the drive itself? Probably not, because most of your attention was focused on your friend. As we get older, we stop relying on the CB because we have built a big store of SB subroutines. If time is what you want to fill then there are a plethora of routine-subroutines to choose from. Which I think is a shame because you are your most human when the CB is in the driving seat.  

When your CB is engaged, time slows down, you become more present, and your subroutines are not filling time up; rather, YOU are. 

(to be continued in a later post)

Interesting reading:

https://www.quantamagazine.org/reasons-revealed-for-the-brains-elastic-sense-of-time-20200924/

https://www.amazon.com/Biology-Belief-10th-Anniversary-Consciousness/dp/140195247X

I have been thinking about people who are 3+ standard deviations out in their respective fields, the people quintessentially associated with the word genius. This includes people like Christopher Nolan (direction), Leo Dicaprio / Marlon Brando / Daniel Day-Lewis (acting), Elon Musk / Issac Newton / Albert Einstein (science), Steve Jobs / Jeff Bezos (visionary), Shakespeare and JRR Tolkien (writers) and Conan O Brien (improv comedy). 

These thoughts popped into my head after watching Tenet, Christopher Nolan's latest movie, which is set in a world where a device that allows you to co-exist in a backward time loop. Objects with their time backward can exist in the forward flowing time universe; this gives them the unique property of having their entropy reversed. Imagine a bullet being un-shot into the gun. The world he sets up makes that possible. After watching the movie, I got a glimpse into Nolan's mind and his capacity to be a "world builder." 

Somehow my mind kept coming back to world-building, and I started wondering whether that is a good measure of a person's multi-disciplinary genius. We all build worlds every day, unconsciously. It is easy to understand if you think about a robot, which is a combination of hardware and software. If it wants to move to achieve its goal of picking up an object at a distance, the sub-steps that happen are:

• Take a snapshot of your world the robot inhabits right now - this means taking in your physical surroundings, the objects in it, how they interact with each other and with you. You also need to keep track of the forces acting on you and how they affect your center of mass.
• Translate the goal to a final world state - what does the snapshot of the final state look like, hold it in memory and optimize decisions that get you closer to the goal world state
• Update your current world state with every step you take until you reach your goal world-state
• A further complication is moving an object to take advantage of it in attaining your goal. These steps change your current world state even if it means you haven't moved (e.g.: moving an obstacle out of the way vs. jumping it) 

We do that every second we act, and humans maintaining balance when we play a sport or walk a tightrope is an advanced application of this. I believe we approach genius with our ability to use a similar muscle to create abstract worlds without a clear goal and interact with them. A director like Nolan imagines the world and creates it on the screen; an actor imagines it and inhabits it; a scientist creates many different versions and sees which obey reality most closely; a visionary sees it and builds towards it; a novelist imagines it and writes it; a comedian invents absurd ones and jokes about their absurdity. 

What I believe differentiates genius is the complexity of the world they create or, in simpler terms, the number of real-world variables they account for in this world. JRR Tolkien's world-building in LOTR is legendary for how varied and detailed it was; Mahatma Gandhi's took into account 600MM citizens of India, a British colonialist empire that relied on "dividing and conquering" and violence to come up with a non-violent movement to achieve independence. Albert Einstein's world accounted for so many variables that he saw through the fabric of the universe itself.

This is not exhaustive, nor a revolutionary idea but I thought useful for recognizing genius - the multivariate nature of their world, and the subtle complexity in it.

(e.g. Genius musicians like Beethoven may not fit, though a possible reach explanation is that they experience a world where their music exists and the emotional impact it has is so visceral that the music flows out through them).  

Everything visible is reflecting light, 

Some manufacture their own, some borrow from the wild

Nevertheless, you catch them all 

All i need to do is point

And you take care of the paint 

Since i was 10 I needed help to paint,

Colors were vivid, but each point corresponded to more than it needed to 

So i used lenses, so I needn't squint nor strain

What started off a window, evolved to become a shield 

A separation between me and everything I saw 

Today I try raising the shield

See what the world looks like, unassisted 

In doing so I open myself to the risk 

Of the window being half open 

or more worryingly, never quite see the same 

I hope I read you shieldless, the other side 

But if not, this was a risk you accepted, and one you will live with 

I have been making conscious early-stage efforts to wrap my head around quantum computing. As Shohini Ghose explains in her TED video, it is a future class of computing, about as different to current computing as a candle was to a light-bulb. Quantum computing emerged because classical computing reached its physical limitations, and further miniaturization was not possible with the same technology. 

A brief expose of my understanding; Quantum computing goes beyond the electron into its subcomponents. Qbits (quantum bits) are these subcomponents and are what enable quantum computing. They have both positive and negative charges (or spins), that is deciphered by using a quantum decoder. These bits are in an eternal "Schrodinger's cat" state, i.e in a superposition of charges. Unlike typical bits that we use to encode information, qbits are never in just one of the charged states; they are in fact somewhere in the middle. Their existence on the spectrum is revealed in probabilities, i.e 80% probability charge 1/2 and 20% is -1/2. Expected value calculations don't mean anything here, but when you choose to "observe" them, they choose a state to show themselves in. The probability density function represents the frequency distribution of their choices. 

Qbits choosing a state is well explained by Heisenberg's uncertainty principle, which says that you cannot observe both location and momentum at the same time, only one. For large objects, both are infinitesimally the same, but observing one facet essentially squashes the other at the quantum scale. If you keep location, you lose the information of velocity and vice versa. Similarly, when you observe qbits they possess both charge states, but observing forces the qbit to manifest one of its states. 

In the 'jerk' of computational evolution, now a single qbit (significantly smaller than an electron already) doesn't hold not only a 0 or a 1, but everything in between. It can encode so much more information, though it does require specialized tech to read the qbit and force redundancy to avoid encoding errors (which are more frequent in quantum computers). So 4 traditional bits can hold one out of sixteen possible states (1/16 of 0000 to 1111) while four qbits can simultaneously hold all sixteen states and choose to show up as one when measured. 

This seems curiously similar to how humans store information associatively rather than deterministically. We may forget a memory because it occurred 16 years ago, but a song or a walk through a familiar neighborhood or a text from someone unlocks that memory and it becomes available. Memory champions use this to their advantage by constructing a quirky tapestry of a story to hold information, and they including sight, smell, sound and sensation to aid retrieval. The more things you associate with the memory, the easier retrieval becomes. 

Could associations act as probabilistic filters to our quantum memory? Maybe our memory is associative and not deterministic because we work at the quantum scale. Quantum computing could unlock huge advances in human cognition. 

Sources:

Qbit diagram - https://www.austinchronicle.com/screens/2019-04-19/quantum-computing-101-a-beginners-guide-to-the-mind-bending-new-technology/