neuroscience is the field which, in the object of its study, has … well, itself! brain that thinks of the brains stands in line with the visualization of the drawing hands' recurrence paradox ¹. just stop here and feel for another minute the mere miracle of having a structure that can be used as a tool for its own inquiry & discovery. one might consider the possibility that our tools are encapsulated in the developed biases towards specific sensory types of information as well as we might hit the boundaries of the system's capabilities (take a look at the first essay discussing the limits of understanding), thus we might just have cracked perception of the actual processes. for example, in the recent Anthropic's report on the Biology of an LLM, there is a fascinating case with an addition - model uses a rough estimation of the addition (18 + 54 should be around 65-75), which then matches with the last digit (8 + 4 → 2) (65-75 + last - 2 → 72). when the model is asked how it does, it proposes a classic explanation with the addition in a column, meaning "… model has a capability which it doesn't have metacognitive insight into. The process by which the model learns to give explanations and the process by which it learns to directly do something are different." ² nevertheless, we have the situation where the brain reveals its secrets to itself; the list might not be incomplete, but here we are to ensure that all the items we can think of and provide relevance are there.

following evolutionary traces, Max Bennett ³ proposes a unifying canvas of the developed features across species. firstly, the ability to simulate a subject's own actions emerged in mammals - sensory neocortex, which enabled model-based experience (aka extracting the essential information from the world to build its corresponding model and use it for internal simulations) together with another structure, agranular prefrontal cortex, which enabled own's agency experience (aka modeling own's actions as well as constructing an intent based on the context from emotions and memories). together, they serve as an interactive imagination system - if I am about to do this thing, what will be the next state of the world? one might find the implementation of this stage of intelligence in the prominent Dreamer model ⁴ in the reinforcement learning field, which uses latent (internal) trajectories after learning compact world model to improve its efficiency while not experiencing these imagined trajectories per se. learning occurs not from the actual experience but from the simulated one.

next, primates were granted the so-called "theory of mind" ability. Sally-Anne test introduces this concept neatly: you are told a story about a girl, her doll, and her mum (modified version from the original one). girl left the doll in her room and went to school, while mum, during that time, took the doll and put it in the washing machine. what place will a girl search for the doll after returning from school? children under roughly 4 years will say that girl will look for the doll in the washing machine while we all understand that girl doesn't have information about its unexpected bath procedure. "theory of mind" proposes the observation that we develop the understanding of "others having distinct experience and information", while without that we would think that everyone possesses the same knowledge of the world as me. notice that the evolutionary leap discussed in the previous paragraph doesn't require self-reference, it just needs the ability to render one's own experience into the future, without particular distinction is that me or is that someone else. "theory of mind" ability requires distinction between me and you, and this guy over the corner: everyone possesses different knowledge. in this moment, we are able to learn not only from our actions (real and imagined) but also from real actions others took by observing them.

finally, we arrive at the unique proposition that human intelligence holds. thanks to the ability to communicate, we are able not only to observe the actions of others but also actively transfer them as if we are observing them (of course, with data compression through the language channel). still, most likely, you won't have "pure experience" as it was during the interaction, it will already be propagated through the narrator's context, which is being constructed as one perceives it. thus, in the very end, we do have modified (including imagined) actions from others, we are learning by the imagination other intelligent systems (other humans) hold. one image is worths a thousand words or one life, a million lives, thanks to interaction and imagination! plurality of knowledge gained us the opportunity to learn from others which is, let's agree, a superpower - you don't need to experience the thing yourself (up to some point, as the procedural knowledge still has distinct neural substrates from declarative) as well as you don't need to be someone else (aka having change your nervous system instantly) to imagine course of action differently as someone else - you just need listen to them.

combined with the feeling of a particularly strong interconnectedness and our age of blurred social class where the expected high margin of the achievable has, as such, disappeared, we become vulnerable to our own superpower. we experience life-changing stories with every visual stimulus in reels and tiktok, as someone could achieve "everything from nothing", and it's easy to fall into the feeling that if you are not there (somewhere above), you failed (we see our achievements as an "expression of our uniqueness"), thus we experience normality as personal failure ⁵ as well as it deprives the feeling of the actual life itself. I don't personally have a working proposition on the workaround except remembering that whatever doesn't help you right now (for example, you caught this feeling in a discussion with friends or by scrolling social media) should be acknowledged, moved away (if you don't have the opportunity to reflect on it) & met in the specifically dedicated time as the part which wants you to experience everything all at once. compassionately accept this desire and let it be for some time, whatever emotions it brings to the ground.

as the last thought, I would like to speculate on what might be the next step in the evolutionary leap. predicting while being unaware of priors (what is the reason for life to be and what it operates on?) is undoubtedly not really a reasonable approach. please be particularly sharp-minded when reading the following part, as it contains speculations! consider the evolution course: learning from own's actual actions (direct reward/punishment) → learning from own's imagined actions (simulation) → learning from other's actual actions (mentalization) → learning from other's imagined actions (communication). the latter is actually the prerequisite for exponential growth - each of us "mines computations" (that being said, interacts with the world by performing actions and learning from them) and extracts the necessities for the joint accumulation. the limit we might hit lies in the inability to "consume" that much of the information by the individual, so my guess is that either pressure from knowledge flood will push existing neural structures to more efficient representation (think of quantum processing of different computational pathways; I am not a trained physicist but well, why don't we are going to see the increase in the scale of quantum effects) or/and the unit will be changed from individual to community (actually, social forces always resembled the ones which take place in the brain. excitatory (pushing) and inhibitory (stopping) signals here and there).