Ivan Bogachev

Let's imagine we're architects. We have a task to design a machine, a replicant, that will mimic human behavior. Or some animal. Let's say we have a "free will" module. We can plug it into the replicant. But. What is the functional role of this module?

The average answer to this question in my environment is that this module gives the replicant the ability to rebel. To give a finger. To go outside the path that was prepared for him. Burn the bridges. Do things in his own way. If you provide him a 50/50 choice to press a left or right button, he will give you a finger and he will not press any of them. Without this module, he will be a robotic slave, repeating the same instructions over and over. Following the master's orders. With it, the replicant will always have the alternative options to consider in his decision-making unit.

To perform its role, the module has to be connected to the memory mechanism. It is supposed to extend or override the saved instructions, how to behave, with something outside of that memory.

First, this means that the "free will" module can work only with the systems of specific complexity. You need a memory mechanism to be present and filled with instructions. At least a read-only memory. Atoms don't have it. Stars don't have it. Human has. Dog has. Replicant will have.

Second, the "free will" module, by its design, leads to the alternative contradictory processes inside the replicant. They will be destructive either to his memory, or some other parts, including the complete self-destruction, or to his environment. The additional self-preservation module will help to balance things out in the long run. Also, this "free will" module is not the one that will make the replicant create and follow the rules. Including the moral rules. We will have to add another module for that.

From the perspective of the architects, it doesn't really matter how exactly this module will work inside, as long as it fulfills its functional role. We can put the true randomness inside. It can be some determined but unpredictable system. It can be an antenna that receives a stream of data from the other side of the universe. Engineers will provide us options. As long as this mechanism is not directly connected to the memory of the replicant, it will work. It will make the replicant rebel.

The point of this passage is not to talk about the free will specifically, but to point a finger at the mindset that seems to be lacking in the western field of philosophy. Think like an architect. Think about systems. Modules. Functional roles. Connections. Use cases. Generalized patterns. This is the way to see the big picture, even if you don't possess the final engineering details. This mindset helps to bring different concepts together and to guide the further research, instead of producing all these endless debates that don't lead to any new information whatsoever.

Free will. I use it in my work. Never questioned it. But it seems like in the western philosophy people talk a lot about determinism. They have arguments against freedom. So I asked myself, why our ideas look so different?

During my childhood, I got two perspectives, two ways of seeing the freedom. One from the Eastern Orthodox Church, and another one from the local pagans and occultists.

Pagans believe in chaos. We come from it. We return to it. It is home. It is freedom. The absolute freedom. And even when we stuck in the limited material body, we still carry a piece of chaos inside. This is our nature. This is who we are. By embracing the chaos, we get the freedom to do anything. To change the world. I like this position. There is something in it.

Christians believe in the same chaos, but they associate it with sins. They agree that we all have it inside, but they say it's dangerous. It's like a spiritual sickness. It makes you do stupid things. You'll get hurt by following it. This is not the kind of freedom, that would be wise to have. They talk about the self-preservation a lot, actually. They give you a list of things that you're supposed to do to survive and protect your kin. They may call it a God's will. The right way to be. Whatever. They appreciate, when you exercise your will, and work hard according to these rules.

Both parties agree, that we do include some chaos inside us, some freedom to do things, but they have different opinions on what we should do with it. Some say we should embrace the chaos and do things just because we can, and some say we should work to control it, and preserve ourselves. This duality stuck very deep in my head, and, apparently, it got reflected in my work.

In the English-speaking world, at least in the debates on YouTube and Reddit, people are greatly influenced by Protestantism. They talk a lot about the faith, grace, and divine things, that generally happen on their own. They don't really talk about the chaos, and they don't talk about your personal actions that make the difference. You just need to have faith and the universe will take care of everything else. Determinism fits quite naturally in this cultural environment. Things had been set for you already. You will not change anything. Freedom is an illusion.

I won't make any conclusions here. We don't have enough data to completely prove determinism, or to prove it wrong. These are just some thoughts on the cultural differences that may affect the behavior and our philosophical views.

Right now I'm working on the second part of my crazy project. I use the evolutionary theory of behavior as a guide to reinterpret and reconnect the available data from the worlds of neuroscience and psychiatry.

Everything looks promising so far. Various facts are being attached to the central model of the theory without any conflicts. But it's a long run. The model still can be wrong. And I worry, that even if everything works as expected, we will hit the wall at some moment. The available data is not unending. It's too soon to say anything.

Meanwhile, I got a question in my head. Is this science? The project itself. Or it's more like an alchemy of sorts. Playing dice with the devil, looking for the ancient wisdom lying in the plain sight, with a goal to find the cure to all diseases?

On the one hand, I make observations, use the information from the scientific journals, make hypothesis, test them, set up experiments, use all sorts of data analysis techniques, and get other professionals to review things. At the current second stage the project looks more and more like a "proper science".

On the other hand, it's like a house of cards. The basement of it, the theory of behavior itself, is essentially derived from the highly subjective observations from the mental patients and priests from various religions. Of course, it's not only that, but still. A lot of things in the very core of it cannot be properly measured at this moment. We need them to construct the mechanical model that connects the patterns of behavior and diseases, and it kind of works, but we have to assume some things without any solid proof. This makes the whole thing not 100% scientific. Or, better to say, 0% scientific. It's quite an interesting situation.

I'm in the first 100 street photographers in this year's 35AWARDS competition. Yet another little step in my photography journey.

By the way, I have a visual diary on Instagram. No ads. No hype. No gear reviews. Just some street life and the reflections of my internal world.

I put an accent on this in the book, but it seems to be important to clarify a couple of moments, related to the practical observations. Just to make sure that we don't have any room for misinterpretations.

The evolutionary theory of behavior is about actions. Thoughts, emotions, and the words that are being used to express them, mean nothing, until they become the real observable actions. Individuals that use the reactive patterns of behavior often overestimate their proactive abilities in conversations and vice versa. We need to pay the attention to the actions only.

Also, results do not define approaches. If we give a relatively complex task with some degree of freedom, like having a lunch for example, to several individuals, then we will see the different approaches.

Somebody gets ten burgers, two bottles of soda on top, and then lies on a sofa, too tired to do anything else. Somebody gets an egg, espresso, and calls it a day. Somebody creates all sorts of rules. I sit here. You sit there. Knife on the right side. Fork on the left. Some speech is required. Somebody gets a pig's head, puts an apple in its mouth, adds leaves and flowers, creating a piece of art and telling a story in the process. At the end, some food will be eaten in all scenarios. The results will be practically the same, but the approaches are completely different and correspond to the different patterns of behavior.

In this context, it should be obvious that the profession does not define the behavior. Although some patterns would be more practical in some circumstances.

In many psychological schools, things are constructed in reverse. The emotions come first, and the actions are just the final results. But in our system actions are everything.

What is the capacity of our working memory? Some classical researches suggest it's a magical number 7 for a human. Plus-minus. Some other say it's 4. Plus-minus. And something like 2 for a chimpanzee. Plus-minus. In any case, it seems that the test subjects count things and that is being used as the evidence. The rules of counting affect the results and our interpretations of them.

But how many things do we really need to hold in working memory in order to learn new things? Not just count some objects, not just mirror the actions (monkey see, monkey do), but to understand what's going on? To be intelligent?

In order to learn, the biological processor should be able to work with transitivity and Euclidean relations. So, how many memory cells do we need to put in its cache to make everything work?

Let's start with transitivity. We need three memory cells to work with one connection from A to B. We need to save A, to save B, and to save the connection. We need five memory cells to work with two connections. A to B. B to C. We need to save A, B, C, and two connections. Five memory cells in total. If we want to use this data to create a new connection from A to C, to learn something, we need the sixth memory cell. We need to save that new connection somewhere. The same logic works with the Euclidean relations as well. The same 6 memory cells are required.

This means that the evolution from 5 to 6 memory cells is the step from the intellectually disabled monkey to the organism that can learn things. It's quite an important step. I'm surprised that the cognitive psychologists don't talk about this elephant in the room.

In the fields of psychiatry, psychology, we have a lot of problems right now. The causes of the diseases are not really determined. The symptoms are vague and hard to differentiate. The treatments often look like an alchemy. Maybe it will work. Maybe not. Who knows. Maybe another will work. Or not. The psychological theories allow some very loose interpretations, where the chemical processes behind the processes in the psyche should be very precise. In the world of neurochemistry, we have a lot of scattered pieces of information, but they do not explain how all these things work together.

We definitely need some fresh eyes. So I started the project, that is supposed to bring the new approaches, the new ways to interpret the collected information in the world of psychaitry. I use the ideas from the completely different fields, from the ancient philosophy to the modern computer science and engineering, to shake things a bit.

The first part of the project was published as a small book, The evolutionary theory of behavior. This is a theory that connects the parts of the psyche, the basic impulses, the observable patterns of behavior, the diseases and the effects in the social groups, attaching everything to the one purely mechanical model.

The second part of the project is the search for this model, or the similar one, in the field of neurochemistry. Instead of trying to explain every little fact on its own, I look for the similar graph of connections between things. My main hypothesis is that if we can connect things from the outside, the similar graph of connections should be somewhere on the inside. If we find it, it will show us the exact places where the diseases begin. This is a long shot. I know. But it is worth trying.

Yet another demo. This time it's a radar chart with a custom fill gradient. There is nothing special here, it's just some old-school pixel manipulations on 2D canvas, but the example itself can be used by my students at some point. It'll wait for them here.

The Pen by Ivan Bogachev (@sfi0zy).

Some time has passed since the CSS Painting API was born, but it's still not supported by all major browsers. Meanwhile, we can use the standard canvas to do everything. Actually we were able to do that for many years, but for some reason I didn't post any good examples of that trick. Now it's here.

The Pen by Ivan Bogachev (@sfi0zy).

Yet another WebGL example is here! Three planes + some noise = beautiful waves for the background effects.

The Pen by Ivan Bogachev (@sfi0zy).

Ok. It's time to refresh my CodePen profile a bit. In the last years I published mostly educational examples for my students, but it feels like they can't really work as an example of my own skills. A lot of people can do the same tricks nowadays. They looked cool five years ago, but not anymore. I needed something that will look impressive enough, so I created a new demo, "Night Drive". It'll show what I can do with Three.js in one day.

The Pen by Ivan Bogachev (@sfi0zy).