BLUE & ORANGE blog_e

neuroscience, consciousness, AI

Consciousness research series: adding chapter 11 and 15


Here consciousness research series: the consciousness research series has been updated. Today shows Chapter 11, 15 adding to Chapter 0, 1, 3, 4, 5, 6, 7, 8, 10.

So far, we have discussed "Qualia" in Chapter 5, "Hard Problem" in Chapter 6, "Consciousness Model" in Chapter 7, "Desire / Purposiveness" in Chapter 8, "Self / Metacognition" in Chapter 10, and so on.

Chapter 11 is 'Semantic understanding / Chinese room / China brain / Time and Place'. Here, I will show what semantic understanding is for humans, other living things, and even LLMs, and what it means for the concept of 'feeling the way it does' of consciousness that follows in later chapters. I will also show distributed semantics, which is important in semantics, as well as symbol grounding for LLMs (ChatGPT) and why the redness of red cannot be perceived (recognized or understood) by LLMs. (This will also touch on consciousness of the Chinese room and the internet).

Chapter 15 is 'Feeling the way it does'. Here, I will show what I consider to be the key to explaining the easy problem side of consciousness, which is important along with the Hard Problem side when presenting the theory of consciousness, by using the 'sense of oneself that it is feeling'. I will also directly answer the question 'What is it like to be a bat?' and show that this part is included in the easy problem side of consciousness and is an independent issue from the Hard Problem.

I hope you will understand that the Hard Problem (see Chapter 6) is a particularly strong contributor to the easy problem of consciousness, to which I will refer in this article, and that consciousness can be established independently on either the Hard Problem side or the easy problem side.

As for other revisions, I have made quite a lot of changes and added three pages in total (with green markings), as I have not been able to fully organize the easy problem perspective, which leads to the consideration of 'feeling the way it does' and their connected semantic understandings.

In this update, I am finding that I actually have not fully explained the sense of agency, and I have mentioned it a little bit on twitter. I will incorporate this into the next update.

In the future, I am organizing the rest of the chapters as concisely as possible to around 20 pages. And I will continue to update the released materials as much as possible with a new revision calendar. I also plan to release the rest of the material as quickly as possible.

Chapter 15: 'Feeling the way it does'




Chapter 11: 'Semantic understanding / Chinese room / China brain / Time and Place'




Chapter 10: 'Self / Metacognition'




Chapter 8: 'Desire / Purposiveness'




Chapter 7: 'Model of consciousness / CBB'




Chapter 6: 'Hard Problem / Easy Problem / Connection between machine and brain'




Chapter 5: 'Qualia / Unit qualia / CBB'




Chapter 4: 'Consciousness, unconsciousness / Monad / Consciousness of stones /Precondition of consciousness'




Chapter 3: 'New ideas for consciousness'




Chapter 1: 'What is consciousness?'




Chapter 0: 'Contents of series of research of consciousness'


Consciousness research series: adding chapter 8 and 10


Here consciousness research series: the consciousness research series has been updated.
Today shows Chapter 8, 10 adding to Chapter 0, 1, 3, 4, 5, 6, 7.

Chapter 8 is 'Desire / Purposiveness'. Here, showing that curiosity and various other desires are materials that enrich the "function" of the easy problem aspect of consciousness. And showing the relationship between Desire/Emotion and Qualia and Hard Problem and showing the basically negative stance on the purposiveness of desires.

And Chapter 10 is 'Self / Metacognition'. Here are detailed descriptions of the self, briefly presented in chapters 5 and 7, focusing on the sense of bodily ownership. And showing that the self is a function of the easy problem aspect of consciousness, touching on the function of consciousness and the definition of meaning understanding.

Also exsiting Chapter 1, 4, 5, 6, 7 have also been slightly corrected.
In particular, in Chapter 7, at the CBB model of consciousness, "Function of Consciousness," p. 21, is added.

The following are the important points 1-6 for understanding consciousness in this slides through science. Please see twees arounf the end of 2022 to the beginning of 2023.
1. There is an affinity to IIT.
2. There are two types of qualia: unit qualia and compound qualia.
3. It's important to think Hard problem is only a narrow part of the scope of consciousness.
4. There is a perspective of dividing Hard Problem.
5. A certain Hard problem is beginning to melt.
6. The idea of CBB is presented for a vast easy problem area within the scope of consciousness.

After this I will release the rest of Consciousness research series slide documents stepwisely. Every chapter will have about 20 pages and there are 18 chapters. It's a plan. So far I wrote all slide documents based on scientific facts basically. I expect a quiet discussion based on this. A trial is to show by colors for comments and unique ideas. Pale orange shows comments by author, and pale blue shows series unique ideas.

Chapter 0: 'Contents of series of research of consciousness'




Chapter 1: 'What is consciousness?'




Chapter 3: 'New ideas for consciousness'




Chapter 4: 'Consciousness, unconsciousness / Monad / Consciousness of stones /Precondition of consciousness'




Chapter 5: 'Qualia / Unit qualia / CBB'




Chapter 6: 'Hard Problem / Easy Problem / Connection between machine and brain'




Chapter 7: 'Model of consciousness / CBB'




Chapter 8: 'Desire / Purposiveness'




Chapter 10: 'Self / Metacognition'


Consciousness research series: adding chapter 6 and 7


Here consciousness research series: the consciousness research series has been updated.
Today shows Chapter 6, 7 adding to Chapter 0, 1, 3, 4, 5.

Chapter 6 is 'Hard Problem / Easy Problem / Connection between machine and brain'. Showing What is Hard Problem, and explaining the reason of Hard Problem with keywords of 'melting Hard Problem' and 'optical illusion-like transformation’, etc.

And Chapter 7 is 'Model of consciousness / CBB'. Showing CBB (CONSCIOUSNESS BUILDING BLOCK concept) as a model of consciousness, & showing functional aspects of consciousness: self, emotion and motivation, semantic understanding, creativity, etc.

Also exsiting Chapter 1, 3, 4, 5 have also been slightly corrected.

The following are examples of preconditions for understanding consciousness through science.
1. There is a consciousness hypothesis that distances itself from dualism, and consciousness is theoretically copyable.
2. Hard Problem is only part of the scope of consciousness.
3. The qualia that the living organism actually feels just happen to be generated by genetic algorithms, etc., and this is not negated by theory.

After this I will release the rest of Consciousness research series slide documents stepwisely. Every chapter will have about 20 pages and there are 18 chapters. It's a plan. So far I wrote all slide documents based on scientific facts basically. I expect a quiet discussion based on this. A trial is to show by colors for comments and unique ideas. Pale orange shows comments by author, and pale blue shows series unique ideas.

Chapter 0: 'Contents of series of research of consciousness'




Chapter 1: 'What is consciousness?'




Chapter 3: 'New ideas for consciousness'




Chapter 4: 'Consciousness, unconsciousness / Monad / Consciousness of stones /Precondition of consciousness'




Chapter 5: 'Qualia / Unit qualia / CBB'




Chapter 6: 'Hard Problem / Easy Problem / Connection between machine and brain'




Chapter 7: 'Model of consciousness / CBB'


Consciousness research series: adding chapter 4 and 5


Here consciousness research series: the consciousness research series has been updated.

Today shows Chapter 4, 5 adding to Chapter 0, 1, 3. Chapter 4 is 'Consciousness, unconsciousness / Monad / Consciousness of stones / Precondition of consciousness' that shows minimum consciousness and other various consciousness and precondition of consciousness.

And Chapter 5 is 'Qualia / Unit qualia / CBB' that shows that shows various 'what it is like’(s) that is thought to be qualia. And that leads to the CBB that is the new theory of consciousness: CONSCIOUSNESS BUILDING BLOCK concept.

Also exsiting Chapter 1 and 3 have also been slightly updated.

After this I will release the rest of Consciousness research series slide documents stepwisely. Every chapter will have about 20 pages and there are 17 chapters. It's a plan. So far I wrote all slide documents based on scientific facts basically. I expect a quiet discussion based on this. A trial is to show by colors for comments and unique ideas. Pale orange shows comments by author, and pale blue shows series unique ideas.

Chapter 0: 'Contents of series of research of consciousness'




Chapter 1: 'What is consciousness?'




Chapter 3: 'New ideas for consciousness'




Chapter 4: 'Consciousness, unconsciousness / Monad / Consciousness of stones /Precondition of consciousness'




Chapter 5: 'Qualia / Unit qualia / CBB'


Consciousness research series: the slide document is out.


Here consciousness research series: the slide document is out.

Today shows Chapter 0, 1, 3. Especially Chapter 3 is 'New ideas for consciousness' that shows whole direction of this slide document of researching consciousness in the future.

After this I will release the rest of Consciousness research series slide documents stepwisely. Every chapters will have within 20 pages and there are 17 chapters. It's a plan. In this Chapter 3, I tried not to write extremely controversial points, but to show materials for quiet discussions.

Chapter 1 is 'What is consciousness?' that shows also possible viewpoints in order for many people to find possible new opportunities for researching consciousness.

Chapter 0: 'Contents of series of research of consciousness'




Chapter 1: 'What is consciousness?'




Chapter 3: 'New ideas for consciousness'


Hard Problem of consciousness, while thinking connection between machine and brain


Background / Introduction
In our country a startup started. That tries to demonstrate machine's consciousness with connection between machine and brain when it becomes to have consciousness. Its purpose seems to feel the machine-side consciousness completely by brain-side without feeling uncomfortable. Here, consciousness information inputting from machine hemisphere to brain hemisphere is researched. Then the assumption that the completion of feeling of the machine-side consciousness is needed. And after feeling the machine-side consciousness completely by brain-side without feeling uncomfortable, the trial is that we can realize machine-side consciousness also as same as organic brain.

While thinking connection between machine and brain, we should consider what signal should be suitable for input information with thinking also the implementation of consciousness for computing. Because of Hard Problem, it's the general recognition that there is no theoretical method to explain a color perception (ex. red for red) with machine at this point. This is Hard Problem: though it may be possible to perceive colors consequently with 3 color cone cells, there is no theoretical method to explain. (Though we can find which neuron are firing while feeling "red" and we can distinguish perceiving "red" cells from other cells with fMRI, there is no theoretical method to explain a color perception (ex. red for red).) I understand Hard Problem is: "human sees the colors of the rainbow with the color sequence many understand. But we don't know the reason of the sequence."

In the meantime, how about considering pressure sensitivity, loudness sensitivity, brightness sensitivity, and shape of monochrome that are shown with stepwisely intensity. Those case should seem to be different from the case of color. Apparently those cases are different from olfactory sensibility, gustatory sensibility as same as color (ex. red for red). And for olfactory sensibility, gustatory sensibility, there is also no theoretical method to explain them. If we can do, it's better for machine to perceive color. (Some may say it must show perception of color to show consciousness of machine, because color is the symbol of qualia.) But as you know, it's hard. I am wondering I can choose pressure sensitivity, loudness sensitivity, brightness sensitivity, and shape of monochrome for studying perceiving qualia, avoiding the problem of color. We can study color as second step.

Until now I tweeted about Hard Problem that came to indicate melting very slightly recently. Regarding pressure sensitivity, loudness sensitivity, brightness sensitivity, and shape of monochrome, while a sensor feel perception of intensity and it has the label of the sensor (place and what it senses), it came to indicate melting very slightly.

Confirmation: cohort B seems to indicate melting Hard Problem.
A. color of visual sensitivity, olfactory sensibility, gustatory sensibility (that isn't shown with stepwisely intensity, but shown with quality of sense)
B. shape of monochrome of visual sensitivity, pressure sensitivity, loudness sensitivity, brightness sensitivity, size, rhythm (that is shown with stepwisely intensity)
C. frequency sensitivity (that is shown with linearity, but may be close to cohort A)

It may need more explanation. Basically A to C are all qualia, so they are still hard. But some of them (cohort B) can be distinguished from other qualia and we can feel perception of its qualia without misunderstanding. That is because the some of the cohort B is linked to sensors (physically?) and because the process of feeling it is within the individual itself.

Introducing examples
For example regarding cohort B, with the condition that conscious-entity understands the sensors (place and what it senses), that conscious-entity distinguishes perception of (ex.) pressure sensitivity from loudness sensitivity, brightness sensitivity or other cohort B's. (The conscious-entity means to understand the sensor's "place and what it senses", because the sensor links to brain physically(?) if it is within individual itself. It means the conscious-entity can feel perception of its qualia without misunderstanding.)

For example regarding brightness sensitivity, the conscious-entity can feel that there is a luminous thing and there is no shield between the luminous thing and the conscious-entity. Regarding pressure sensitivity, while it feels something with pressure sensor on the right shoulder, it feels that something taps on there, and (thinking of outer space) while it feels bending on the neck sensor, it feels its head is left alone due to reaction of acceleration of the upper part of the body. If the information is sent into all over brain, the info isn't felt as appropriate one (ex.) brightness sensitivity as if it is felt as stun gun stimulus. It is very important for the info to be linked to brain within individual itself directly.

If the information is linked to Visual cortex for brightness sensitivity, it should be brightness information. That is unthinkable that an info (ex. brightness sensitivity) is treated as other perception (ex. pressure sensitivity) if the info is linked to Visual cortex.

And let's think of evolution of microbe. At first it is OK to escape to everywhere always while it feels pressure sensitivity, loudness sensitivity, or brightness sensitivity. But after some evolution, the microbe comes to select actions from many alternatives in regard to type of stimulus due to increase the survival rates.

Additionally its process of cohort B (the feeling of intensity) can be used for brightness sensitivity with four sensors that are used for four view angles that are used for upper-right, lower-right, lower-left, and upper-left. And as similar process, and with detailed grid cells we can feel monochrome pixel art. And this feeling of pixel art can also be used for pressure sensitivity, and further, the stimulus to tongue also can show pixel art as similar as brightness sensitivity.

Discussion
In the meantime there is no theoretical method to explain a process of cohort A (ex. red for red) even now. (Hard Problem) though it may be possible to perceive colors consequently with 3 color cone cells, there is no theoretical method to explain. "human sees the colors of the rainbow with the color sequence many understand. But we don't know the reason of the sequence."

In this way perception can be largely grouped into three. And I understand there is a perspective that cohort B seems to indicate melting. Considering Hard Problem, an unconsidered idea (like Pretty Hard Problem) doesn't lead a link to next idea. But I think the process that shows analytical ideas from already-known ideas lead links to next ideas.

While considering inputting a piece of consciousness to brain from machine, it means consciousness is completed within the machine (*1). But it's very difficult for machine to say "I felt it." if the target is from cohort A (ex. red for red). (The machine can understand its frequency of the color. But it doesn't mean the machine internal is filled with red color.) In the meantime I think it's rather easier that for the machine to say "I felt it." and seems close to self‐evident if the target is from cohort B (ex. pressure sensitivity). But Its precondition is the process of feeling it is within the individual itself.

So if we try to input this (cohort B) information to internet, tagging (the place the sensor located, and what it senses) is necessary. And if its output is into a living body, output to suitable sensory cortex is necessary. And if its output is into a machine, it is necessary to decode with that tagging.

There may be similar prior research, because this can be tried with thought experiment even in the mind in the past even though they can't think of consciousness implementation to machine then. But after considering this topic, there may be also some hints of cohort A. Some research of "eye" may cause some opportunities.

I didn't hear that some disease cause a malfunction of the colors sequence of the rainbow, (though I didn't investigate all documents). And I am wondering living being under visible rays can see the colors sequence of the rainbow (except infrared ray and ultraviolet ray) if it has 3 color cone cells. What should we think next? Is it same about aliens? There are many viewpoints.

Note: I heard that color impaired don't see a certain color frequency band, and I think it doesn't cause the colors sequence change, though some part of sensitivity of frequency is impaired and that may cause difficulty of distinction (ex. between red and green). Please note I don't have detailed knowledge of disease examples. There may be some hints in searching them. For example, a decline to limit of sensitivity of one cone cell of 3 color cone cells might cause another unknown color perception, and so on.

Next steps
That's it. This is based on the thought of minimum and / or lower order consciousness. (*1) is also so. In this background there are also the thought of IIT, unit qualia, and consciousness building block concept. (Except IIT, it is a little difficult. See link: With Chinese room metaphor it's start point to thinking about CONSCIOUSNESS BUILDING BLOCK CONCEPT
mambo-bab.hatenadiary.jp
This consideration started from the idea that avoid the difficulty of cohort A. And it caused this consideration of Hard Problem consequently. At first I started this consideration with an idea of connection between machine and brain from one startup. I believe this kind of viewpoint studies contribute progress of various research. I appreciate this kind of proposal. I myself understand that there were some progress of the understanding of Hard Problem. And I'm going to outreach this consideration as well as other research of consciousness continuously.

'Self and others' as higher order consciousness

Self and others:
While one's hand touches something, 'what it is touched' can sense 'touched' at the same time. And we can discriminate self ('what it is touched' is a part of 'the one who discriminates self') from others ('what it is touched' are usual things).

'The one who discriminates self' is 'what has consciousness'. Usually even if the one's hand touches something, all 'what it is touched' are others. There should be a time when 'what has consciousness' find self. An infant finds 'what it is touched' = self on one occasion. The start point of finding self is to find unusual (finding 'what it is touched' at the same time) from usual (others).

This 'primitive self': ['what it is touched' is a part of 'the one who discriminates self'] may be researched in the past. So it may be like reinventing the wheel. But do many researchers understand 'higher order self' is based on understanding 'higher order others' (human like others), and understanding the existence of self which has similar trait as 'higher order others'?

In the meantime, I've heard of metacognition of AI. And I heard primitive metacognition is shown as it can sense whether the switch is on or off of its own circuit. But I think 'higher order self' contributes to higher order metacognition. I think if it doesn't show self, it can't show higher order metacognition.

This 'higher order self' should be evoked naturally in a living body. But with von Neumann architecture AI to model self, and for many researchers to understand, an artificial program might be work better.
I show the suitable order for programming (for self and others).
(0. 'primitive others')
1. 'primitive self'
2. 'higher order others'
3. 'higher order self'

'Primitive self' can be implemented with hand model and a sensor for 'what the hand is touched'. The next 'higher order others' is more difficult, because it needs understanding that some of the others work with some logics. 'Higher order self' is far more difficult, because it needs understanding that the 'higher order self' has similar trait as 'higher order others'. But it is 'CONSCIOUSNESS BUILDING BLOCK CONCEPT': http://mambo-bab.hatenadiary.jp/entry/2018/05/07/200250

It's useful for computer programming of consciousness to show self and others. If it is a living body (that is a general purpose program (metaphor) that simulate all of the world), it takes hundreds of millions of years to show self and others. It's not bad to write computer program for experiment to make this hypothesis understood by many researchers. Some special program would work (ex. demonstrating only self and others without requiring natural conversation).

At #ASSC there seemed a discussion topic of 'no self, no fear'. It seemed too naive, though I didn't hear in detail. As you can see from 6/25/2018 tweet, primitive fear doesn't probably need even primitive self, as same as primitive anger doesn't need primitive self.

Meantime higher order fear is probably affected by various higher order self, as same as higher order anger is affected by various higher order self.

There is self as one of higher order consciousness. In detail there are from primitive self to higher order self stepwisely. That is one of 'CONSCIOUSNESS BUILDING BLOCK CONCEPT'.

This entry is retouched based on the twitter on 7/6/2018 - 7/7/2018.