The question we have been asking for centuries, namely how the mind works, is an absolutely fascinating one. Isn’t it?! However, it is interesting that the vast majority of researchers seem to be strategically ignoring it. Even the majority of those researchers who are looking at AI technologies. This possibly in the hope that Artificial General Intelligence (AGI) could be achieved even without answering this question clearly, through an ever increasing repertoire of tricks (Deutsch, 2012). My contention, however, is that there is no way around being able to explain intelligence and mind if we are to program intelligence.

Let me make a daring attempt to at least provoke some thought. What if the mind creates new knowledge through the neo-Darwinian evolution of self-replicating ideas? The idea for this builds on Karl Popper’s conjecture that knowledge evolves by alternating conjecture and criticism (1983). However, it considers this a rather high-level explanation of a richer underlying phenomenon. The idea of this resolves several open problems in Popper’s epistemology. But also, for instance, questions about how humans evolved, how memory works, what it means to believe one thing and not another, and why we are aware of some phenomena and not others. It also confirms and strengthens Popper’s notion that the central thing about the mind, the very thing that intelligence researchers have studiously ignored for the last 70 years, is the generation of knowledge.

Neo-Darwinism, in general, begins with a pretty simple core concept. Namely, replication. Yet the concept is rich in implications.

The first implication of the concept is that if you start with a single self-replicating something, you will pretty quickly have a population of it. And that’s completely independent of what it is. It doesn’t necessarily have to be biological. For simplicity, we’ll call that something an instance of a replicator in this context. That single instance can generate another one that does the same thing, and the first instance can probably generate another one over and over again. And so can the others. Thus, a single instance results in a second, then four, eight and so on.

The second is that such an instance will eventually make a mistake during replication. For it seems all clear to us by now that the world is messy and mistakes seem inevitable. When such an error happens, it correspondingly changes the information that the replica contains compared to the original. This change, which we understand as mutation, is inherited by all its descendants and can occur randomly. At least this encoding is inherited if the instance is still able to replicate. The consequence is that such mutations introduce variations. In other words, a mutated instance is called a variant. It is important to note that in most cases a mutation hinders the ability of a replicator to propagate. But why is this the case? Simple. Because the replicator is already adapted to spread, meaning that any slight change could reduce its ability to spread (see Paley’s comments on adaptation, 1813). A very good replicator prevents this type of mutation from occurring.

Sometimes, however, a mistake can be beneficial. Namely, for the replicator’s ability to spread. That is its main concern. Thus, evolution does not optimize anything per se. It merely favors good replicators. Nothing more, nothing less. In biological evolution, that replicator is the gene, not an organism or group of organisms (Dawkins, 2016).

The third implication is that such better replicators, the very ones that promote replication, will spread through the population. We observe this not only in biology, but also in society (cue memes and other ideas of the mind). More on this later. But now back to the third implication. If better replicators spread out in the population, it means that worse replicators will be left behind. These differences in replication rate are called selection (Dawkins, 2016). Selection can result from competition, predation, mutations, changes in the environment, and so on.

The fourth implication is that complex adaptations can arise because of this repetitive interplay of variation and selection. Adaptations whereby replicators can acquire capabilities that go beyond mere replication. Each of these adaptations is always in service of the primary purpose: replication. For otherwise it would not have been selected. Yet, customizations are sometimes complex enough to have sub-purposes that detract from that primary purpose.

But what does all this have to do with the mind and our actual subject of this article? Popper discovered that human knowledge evolves (1983). So evolution as a phenomenon seems to not be limited to the biosphere, but it is actually also able to explain how we humans create new knowledge. Dawkins later added that evolution also explains how ideas spread between people and change in the process. He calls such ideas memes (meme culture, though probably limited for many in understanding to somehow funny content, is really coming to life on social media platforms right now). Popper offered a rigorous analogy between genetic mutations and conjecture on the one hand and natural selection and criticism on the other. His explanation was a major epistemological breakthrough. Therefore, Popper can be seen as one of the most important AGI researchers of his time.

In his problem-oriented epistemology, Popper argued that we create knowledge by conjecturing solutions to problems, then criticizing them, and finally tentatively accepting them when we cannot criticize them. He viewed the pursuit of certainty and the notion that we should try to confirm our theories as justification error. While these things can make sense, depending on your point of view, there are still outstanding problems with Popper’s epistemology.

Where, for example, do the conjectures referred to come from? They have the appearance of design, and so we must explain their emergence as the result of evolution. The claim that human knowledge evolves through conjecture and criticism is therefore limited in its ability to explain the phenomenon. For conjectures are themselves a product of evolution, which takes place in a mind. By the way, the same is true for criticism. It is also a conjecture and has the appearance of design.

Another problem is that Popper’s evolutionary model lacks the neo-Darwinian notion of a replicator. The reason for this may be that he started developing his theories before the modern synthesis of neo-Darwinism was completed and also before the discovery of DNA. This discovery was the first of a physical instance of a replicator, although of course it was known from theory before. That’s why it was not until the modern synthesis that the importance of replication was generally recognized. Now you may have just caught me flat out. For admittedly: I have only read some of Popper’s books. But nowhere in them have I found any evidence that he included or built replication into his theory of human knowledge. Please correct me if this is a completely erroneous assumption.

Then there are other problems which are not directly related to Popper’s epistemology. Nevertheless, they are interesting and seem to me to need to be solved. Some of these problems relate to memory. Why do we remember some things and not or no longer remember some other things? Why do some of our friends remember some things better than we do? More generally, why do some memories last longer than others? And adjacent to that a very interesting and central question: Why do some people believe some things more than others? Or why does it take more persuasion for certain people than others? But beware… I’ve brought persuasion into discussion. Persuasion is something that Popper’s epistemology does not account for. So you can take that as a personal matter. Because it really puzzles me personally why some people are better or worse with certain ideas.

Well… How can we tackle these problems then? I propose the following conjecture: The mind contains a kind of arena of self-replicating ideas.

When we hear replicating ideas, we may immediately think of memes (Dawkins, 2016). Fair enough. But those don’t really correspond to what I mean by that. For memes are ideas that spread between people. But I am limiting myself here only to the dynamics within a single mind. Accordingly, the context is completely different. The neo-Darwinian theory of mind is not about the evolution of memes. This is despite the fact that, like any theory of creativity, it lays the groundwork for explaining how complex memes can spread between people in the first place. But let’s try our hand at a simple start. The simple concept of replication.

To begin with, a self-replicating idea will generate a set of ideas in a mind. a “pool of ideas.” Since the mind is also messy and mistakes are inevitable, sooner or later one of the ideas will make a mistake during replication. At that moment, a mutation occurs. This mutation thus results in a variant idea. Usually, the variant will have a harder time spreading than the original idea. But now and then such a variant can spread better.

Ideas that spread better will spread in the set of ideas of the mind. And then, again, the selection process comes into play. Because some of the ideas are better suited to spread than others.

Second, because the mind is messy and mistakes are inevitable, sooner or later, an idea will make a mistake during replication: a mutation occurs. This mutation results in a variant idea. Usually, the variant will have a harder time spreading than the original. But now and then, the variant is better at spreading.

Complex adaptations can emerge because of the repetitive interplay of variation and selection. Ideas can thus acquire abilities that go beyond mere replication. Such adaptations, however, do not really have to do anything good to the mind itself. The focus is merely on producing better replicators.

Now this all sounds pretty simple. And somehow I can well imagine that this attempt at explanation comes pretty close to the reality. Better… We can now also address the problems mentioned before. Conjectures are newly evolved variations of ideas. Some things are retrievable in our memory because they are better replicators than others. Memories are not a special kind of idea. They are simply self-replicating ideas of high longevity and relatively high stability. In other words, they manage to remain in memory longer than most other ideas and exhibit higher information stability.

But why do some people remember some things better than you do, for example? Let’s assume a married couple generally remembers their last vacation together. But the spouse remembers many more details. Why? Simple. Because her mind happens to be a suitable environment for the ideas that encode those details. This also favors the propagation of those ideas. In the case of the spouse, this environment seems to have been less than ideal, relatively speaking.

Because some replicators are more durable than others, it can be explained that some memories are present longer than others. When your grandparents tell about memories from their childhood, however, idea instances are not as old as the idea itself, but possibly only a few seconds. At least that’s a very vague guess I have.

Let’s move on to the questions that are particularly exciting for me: Why do some people believe some things more than others? And why is it easier in some cases to convince people of idea x than others? The answers seem soberingly simple. The answer to the first question is because some ideas spread better in their minds than others. Answer number 2: Since in the mind of one person the idea x can spread more easily than in the mind of another.

You see, the answers about memory are very close to belief. That’s because memories are just long held beliefs. Put another way, both are long-lived, self-replicating ideas. Even though there are significant differences between biological and mental evolution, the neo-Darwinian theory of mind in general seems to have a strong unifying character.

The vast majority of mental activity seems to occur unconsciously. We are not really aware of the millions and millions of variants of self-replicating ideas in our heads. These ideas partly work together, others compete directly with each other for propagation in the mind. In both cases, however, we are just, as mentioned, unaware of the replication of the ideas. But we know from theory that such processes take place.

This brings us to the question of consciousness. To possibly one of the most fascinating and mysterious phenomena in the universe to date. Popper (1983) has an exciting conjecture here. Namely that consciousness has to do with disappointments. In other words with expectations. For example, if we expect that if we do x, y will happen, but this y does not occur, we become immediately aware of this non-fact. We could thus say that consciousness seems to have something to do with error correction. We observe this also with children, for example when they learn to walk. When a child learns to walk for the first time, the motor task can be overwhelming. Overwhelming in this example is to be understood proverbially. After all, it often happens that a child falls down. However, in the process, the child’s mind is constantly developing new ideas about how to approach the task. Most of them will not work at the beginning. But some of them will. And so these variations ensure that the child’s walking skills improve over time and become ingrained in memory. Over many retries and associated iterations, the child’s ideas improve, and while the mind does error correction, the task of walking and the associated procedures become less and less conscious. After the child has taken thousands of steps, the child per se will not even know how exactly it is doing so. In detail. It just does it.

It could be that disappointed expectations will play an essential role in any good coming explanation of consciousness. And yet they seem to me to be insufficient. Yes possibly even not really necessary. Think about it. If you sit down, close your eyes, meditate for a while, and observe your thought processes… What thoughts go through your mind? I contend very few will have to do with disappointed expectations. Some of them are just memories or other chatter of your mind. However, we know that they are all ideas. And they all have the appearance of some sort of design. Perhaps part of the mind is like a bubbling cooking water of ideas, some of which manage to bubble up here and there. These could be our conscious thoughts. When competition between ideas is particularly high and the water boils, many ideas bubble to the top where urgent attention is demanded. Consequently, these are our thoughts that occupy us. Perhaps consciousness takes snapshots of the boiling water to more closely analyze ideas that have bubbled up far enough. Or maybe it only looks at one instance of an idea at a time. That could be why we are not aware of replication.

The theory is of course totally incomplete. It does not explain consciousness. We should therefore address this problem next. What if consciousness is software that can be written just like any other software, and that it could run on any existing universal computer? Whatever the case, we will need the neo-Darwinian theory of mind or something else comparable to explain mind and build something like AGI.

Only time and rigorous criticism will tell if there could be any truth to this theory. I am curious and look forward to seeing what happens next.