The next thing I knew, I was wandering around the city at night.
I suppose I became a zombie temporarily from being bitten by a zombie, right?
But I came back to being a human and lay on my bed.
This was how much I could recall, I couldn’t recall what had happened afterwards.
Most probably, after having returned to be a human, I went back to be a zombie again, and wandered around the city.

The last time I came back to be a human, my body parts ached and I could barely walk.
Because rigor mortis had set in, it hurt very much when I bent the joints of my limbs.
I came to realize that if I walk with my arms extended forward, an effort not to bend my joints as much as possible, I can walk properly.
I was strangely convinced that “hey, this is the so-called zombie walk”.
Without realizing, I had learned the zombie walk.
By the way, zombies don’t have qualia, right? So even without qualia, they can learn?

I explained about how the brain of a zombie has no stream of symbolization（ventral stream）but only the stream of action（dorsal stream）in the last post.
The stream of action is directly linked to the real three-dimensional world.

When you ride a bicycle for the first time, you can’t keep the balance right.
After several practices, you start to learn how to pick yourself up from falling off, how to swerve the handles and how to pedal – the ways not to fall off from the bike gradually.
This is learned in the stream of action.

The action of swerving the handles is done unconsciously.
If you are thinking “if I’m falling on the left, I should swerve the handles to the left” in your head, the next think you know: you are falling because it took too much time thinking.
When you think, things are symbolized and your consciousness control those symbols.
Symbols that your consciousness identifies are qualia.
Qualia is identified as “this is it” by your consciousness and can be controlled by it.

The moment your consciousness identifies something, it become a symbol disconnected from the real world. That’s why it is controllable apart from the real world.
It means that you can think about swerving the handles to the right, to the left without actually taking the actions.
However, while you are thinking in your head, the real world is changing and your bike is falling.
Symbolization takes too much time.

If a frog thinks “how should I catch this fly?”, the fly will go away.
The frog must extend its tongue the moment it identifies the fly; or else, it wouldn’t catch the fly.
There’s a need to take actions in order to adapt to the ever-changing real world.
This is what I mean by, it is directly linked to the real world.
In order to survive in the nature, qualia is unnecessary.

One will learn to ride a bicycle without qualia.
One will be able to adjust the timing of actions so that it matches the real world appropriately.
So one can learn without qualia.
Even if one becomes a zombie and the body sets in rigor mortis, they will learn to keep balance to walk.
I’m a bit relieved now.
（Haha why are you relieved for this?）

 

AI is very good at optimizing adjusting body movements.
Deep learning will be able to optimize automatic driving of cars or automatic control of drones.
Please note that deep learning is only capable of optimizing actions in the real world.
AI which is made up of deep learning only is like a zombie – AI without consciousness.

Let’s think a little bit more about how zombies or animals that don’t possess consciousness like humans see the world.

Animals run after their preys once they identify one.
When they identify a predator, they will run away or hide from it.
So they will only take patterned actions according to the information perceived.

They are not capable of thinking like “I wonder what there are behind that mountain?”.
To be able to think like this, they would need to have a sense that they exist in the three-dimensional world.
To be able to distinguish the world from oneself, one needs to be able to symbolize the world and oneself in the first place.
Animals without the stream of symbolization do not perceive the world in such ways.

I wonder what kind of a world it is, which is identified only by the stream of action and where the stream of symbolization does not exist.
Imagine that sensation you feel when you are riding on a bike while keeping the balance.
However, you may not think: “I can reach the post office if I go straight on the road”.
Since this thought can only be generated by the processing of the stream of symbolization, animals cannot have it.

Concentrate on the action to ride a bike while keeping the balance.
If a cat suddenly appears from the side alley, you just swerve the handles quickly.
This is what animals feel, which do not possess consciousness.

Just like humans, they see the real world with their eyes.
However, the sense of “I am riding on a road” is only possible when one identifies oneself in the three-dimensional world; in other words, only when one is capable of symbolization.
Symbolization involves rebuilding of what exist in the three-dimensional world.
To build the three-dimensional world is to set a platform of the three-dimensional world in your head and to place things you see there.
A three-dimensional platform is like a coordinate space where you can place 3D models. To replicate the real world, you place 3D models of the road, trees and electric poles.
While you are on a bike running, you may feel the scenery of electric poles and trees passing by; this is because you are rebuilding electric poles and trees in another world in your head.
Of course, in such virtual world includes yourself.
The feeling of yourself existing in the world comes from the fact that your consciousness identify the three-dimensional platform.

Your eyes are only capable of capturing part of what lies in front of you.
Think of your vision as a circle that is 10 cm diameter big.
Aside from that, you are not really seeing things properly.
Nonetheless, you clearly feel that the world exists beyond what you see before your eyes.
You don’t feel that the world besides the part your eyes see does not exist.
Like this example, the reason why you feel this presence of existence is because you rebuild what you see in a virtual world and your consciousness identifies them, instead of directly identifying what your eyes see.

Then, why did humans create this complicated system in which one needs to build a virtual world and identifies it?
That is because it allows us to perceive the real world more accurately.
Even when you are not seeing, the world continues to exist.
If the world cease to exist while you are not seeing, that world would be far detatched from the reality.

Yet, one can live in the world by directly seeing things and responding to what one feels.
This is how species except us humans survive.
However, animals do not feel as though the world that cannot be seen cease to exist.
They don’t even have the sense that they exist in the three-dimensional world.
They live in this world with only the kind of sense you’d feel when riding a bike.
When riding a bike, you can keep pedaling and not falling, without identifying the three-dimensional space or yourself.
You may only respond to the perceived real world.

We can create a robot that can ride a bike without falling.
It would be able to avoid a cat that suddenly appears from the side alley.
Seen from the outside, there would be no difference between a human riding a bike and the robot riding a bike.
Yet, the kind of world that robot sees and feels is completely different from the one we humans see and feel.

In the near future, we’d expect to see AI robots that look exactly like humans, or automatic driving cars that drive much better than humans.
That doesn’t mean such AI will identify the world just like humans do.
It will be very challenging to create AI that can communicate with humans equally, unless we work on things like how it identifies the world to the same degree we humans do besides its appearance and movements.