How do atoms build thoughts and emotions?

[A.T.]

No it isn't objective proof that we have the same subjective perception, because our brains are not identical

Suppose that a complete theory of the mind did include some unmeasurable subjective aspects of consciousness. Suppose further that they serve as postulates from which the measurable aspects were derived. Then you might indeed consider such evidence to be objective proof of the subjective experience.

In the text you quoted I was talking about comparing subjective experiences of two different people. But you seem to talk about proof that subjective experiences exist.

 

[A.T.]

This is a matter of language, not of "different aspects of the same coin". In the German language, what PeroK wants to talk about is called "Selbstbewustsein".

The most specific term for subjective experiences I could find is: "apperception" (German: "Apperzeption"):
https://en.wikipedia.org/wiki/Apperception
https://de.wikipedia.org/wiki/Apperzeption
And it seems to be generally regarded as a part of "consciousness" (German: "Bewusstsein"):

While @PeroK seems to be trying to reduce "consciousness" to "introspection" (German: "Selbstbeobachtung"):
https://en.wikipedia.org/wiki/Introspection
https://de.wikipedia.org/wiki/Selbstbeobachtung
The term for it you mention: "Selbstbewustsein" might mean the same in academia, but is colloquially also used for "self-esteem" (German: "Selbstwert").

Anyway, I don't see the point in this reduction of the definition of "consciousness", because as I already said: It doesn't solve the "hard problem of consciousness". It just becomes the "hard problem of apperception".

 

[Dale]

In the text you quoted I was talking about comparing subjective experiences of two different people. But you seem to talk about proof that subjective experiences exist.

Both. If the model derived the measurable mind/brain state from a model of the unmeasurable subjective experience, and if you had two different people whose measurable state indicated the same unmeasurable experience, then you would indeed take that as objective evidence that they were having the same subjective experience.

 

[A.T.]

Both. If the model derived the measurable mind/brain state from a model of the unmeasurable subjective experience, and if you had two different people whose measurable state indicated the same unmeasurable experience, then you would indeed take that as objective evidence that they were having the same subjective experience.

Only if I trusted the model. But how would you ever validate such a model?

 

[Dale]

Only if I trusted the model. But how would you ever validate such a model?

The same way you validate any scientific model: by experiments.

 

[A.T.]

The same way you validate any scientific model: by experiments.

Describe the experiment you would do here.

 

[dextercioby]

Sorry for the interruption, but I marvel that one would ask a physicist: "What is conscience, what are thoughts, and emotions for a physicist?", when even a much more basic question is not answered (or maybe it is, but I am ignorant): "When do(es) a large collection of huge organic molecules become a living organism, in other words what is life for according to a physicist or a quantum chemist?"

 

[Zaciekawiony]

Answering such questions is difficult due to the ambiguous nature of the concepts involved. Our descriptions of self-awareness are primarily phenomenological, whereas descriptions of "thinking" or "emotion" rely on the single case accessible to us - human experience - even though similar processes undoubtedly occur in animals. We need generalizations that describe effects associated with the operation of complex nervous systems, within which what we call consciousness represents a special case.

The most common hypotheses contain significant gaps and lack precision. Discussion frequently centers on emergent processes—the appearance of new phenomena resulting from the activity of complex systems, phenomena not present in the individual constituent elements. The behavior of carbon atoms does not encode information about the appearance of a diamond, as certain optical properties only acquire physical meaning once the macroscopic phase has formed. Nor does the diamond’s existence inherently contain information about its market price. A similar logic applies to the conjecture that cognitive functions emerge from the complexity of interconnected neuronal systems: while consciousness and thought are among these cognitive functions, they cannot be deduced from the workings of a single cell. Ultimately, this amounts to an answer of "it happens somehow."

 

[gentzen]

"When do(es) a large collection of huge organic molecules become a living organism, in other words what is life for according to a physicist or a quantum chemist?"

You know https://en.wikipedia.org/wiki/Ernst_Peter_Fischer ?
He will love to tell you the story of Niels Bohr urging physicists to venture into biology, Max Dellbrück taking up the challenge, Erwin Schrödinger popularizing Dellbrück‘s model in his book „What is life“, which later motivated Watson and Crick to propose the double helix.

I think others were motivated too to dive into those questions, like Heisenberg, but I only remember Schrödinger in sufficient detail. Life has to be able to store information and later read it again and interpret it correctly in its context. Therefore a virus is „adjacent“ to life, but not really alive. And todays LLMs are closer to a virus than to being alive or conscious, because they still have to rely too much on humans and our culture for storing and recalling information.

And as long as an unborn baby is not yet able to store its experiences in a form that it can later recall and interpret correctly, it is not yet fully self-aware.

 

[Dale]

Describe the experiment you would do here.

Surveys, EEG, fMRI, whatever measurable results the model predicts. Just the same as with any other scientific theory.

I am not suggesting anything unusual here. I am just suggesting that a hypothetical scientific theory of the mind should be treated the same as any other scientific theory.

 

[A.T.]

If the model derived the measurable mind/brain state from a model of the unmeasurable subjective experience, and if you had two different people whose measurable state indicated the same unmeasurable experience, then you would indeed take that as objective evidence that they were having the same subjective experience.

Only if I trusted the model. But how would you ever validate such a model?

Surveys, EEG, fMRI, whatever measurable results the model predicts.

What would you compare with what to validate that the "derivation" of measurable from unmeasurable by the model is correct?

 

[Herman Trivilino]

For example, I love my wife and I love my job. And the love I feel for my wife is greater than the love I feel for my job. So it seems possible that love could be quantified.

Perhaps. But isn't there more to it than magnitude? The way I love a significant other is different from the way I love my children, or siblings, or parents, etc. And the way I love people is different from the way I love things.

 

[Dale]

What would you compare with what to validate that the "derivation" of measurable from unmeasurable by the model is correct?

For that you would have to look at the particulars of the derivation itself. Just like any other theory.

Since this is a hypothetical theory the hypothetical derivation is hypothetically correct.

Again, this is perfectly standard and what is already done with mainstream physics theories. You take an unmeasurable postulate, like the one way speed of light being $c$, and you derive the testable consequences. You make measurements and if they agree then you can take that as evidence supporting the unmeasurable postulate. All while being aware that the postulate itself is unmeasurable. And that is all OK.

 

[A.T.]

Again, this is perfectly standard and what is already done with mainstream physics theories. You take an unmeasurable postulate, like the one way speed of light being c, and you derive the testable consequences. You make measurements and if they agree then you can take that as evidence supporting the unmeasurable postulate. All while being aware that the postulate itself is unmeasurable.

The first issue I see here is: In what language would you formulate those postulates about subjective experiences? The main problem with comparing subjective experiences is from the inability to objectively describe them.

But assuming you solve the above issue somehow, the next issue is: The subjective experiences might not be objectively measurable, but they are subjectively observable by everybody. So you are not postulating something that nobody can probe directly. What if people disagree with what you postulated about their subjective experiences, even though you derived the correct objective brain states from your postulates?

 

[Dale]

The subjective experiences might not be objectively measurable, but they are subjectively observable by everybody. So you are not postulating something that nobody can probe directly

I don’t think this matters scientifically. If they can probe it in a way that could falsify the theory then it is measurable (and hence not part of the “subjective experience” as you mean it). If it is not measurable then whatever “subjective probing” is done cannot falsify the theory. So scientifically it is moot.

What if people disagree with what you postulated about their subjective experiences

Then they would not accept the theory. This is exactly what happens with theories. Historically, not everyone accepts any given theory. But if experiments continually validate it, then the community as a whole accepts it and moves on.

 

[256bits]

And as long as an unborn baby is not yet able to store its experiences in a form that it can later recall and interpret correctly, it is not yet fully self-aware.

A great sentence!
For an odd 70 years ( life span ) someone is presumed 'conscious'. At the end, does the 'consciousness' vanish, or continue in some other state not attached to physical mass, other than in the memory of other select 'conscious' humans. For animals and other life forms with lessor quality of consciousness, is it not, or should it not be, of the same aspect, or is there only at when a certain level is crossed that the situation does apply?

The assembly of atoms has to explain,
- a return to ourselves when we awake after being 'unconscious' ( sleep, coma, ... )
- a return to ourselves after induced 'unconscious; state ( chloroform, ... )
- degenerate disease that effects the mind ( dementia, ... )
- mind altering drugs that affect personality ( alcohol, LSD, ... )
- traumatic experience affecting personal outlook and response ( blow to skull, PTSD, stroke, ... )
- sex differences of the mind ( male, female, ... )
- level of maturity ( nature vs nurture ie learned or reflex or perception )
- etc

I am sure there are others situations needed to be explained for other than a healthy state.
With a mix of the colloquial consciousness ( Lawyer: Was the deceased still living when you arrived at the scene, and in a state of being conscious or unconscious? ), and some other definition, the ability for the speaker and the audience to be on the same wavelength as to what is being discussed can be stressed.

 

[A.T.]

I don’t think this matters scientifically. If they can probe it in a way that could falsify the theory then it is measurable (and hence not part of the “subjective experience” as you mean it). If it is not measurable then whatever “subjective probing” is done cannot falsify the theory. So scientifically it is moot.

So what the people actually experience subjectively doesn't matter at all in that model?

If the model derived the measurable mind/brain state from a model of the unmeasurable subjective experience, and if you had two different people whose measurable state indicated the same unmeasurable experience, then you would indeed take that as objective evidence that they were having the same subjective experience.

I don't see how you can infer anything about the actual subjective experiences from a model in which the actual subjective experiences are irrelevant.

As far I can see, all you have done is introducing an extra layer between objective measurements and subjective experiences. But that extra layer isn't related to the subjective experiences any more strongly than the objective measurements, so I don't really see the point of it.

 

[Dale]

So what the people actually experience subjectively doesn't matter at all in that model?

It might matter in the model, but not scientifically. It would be a part of the model that cannot be experimentally verified or falsified. It would be similar, in that respect, to the one way speed of light in special relativity.

I don't see how you can infer anything about the actual subjective experiences from a model in which the actual subjective experiences are irrelevant.

Same as you would with the one way speed of light in relativity.

As far I can see, all you have done is introducing an extra layer between objective measurements and subjective experiences. But that extra layer isn't related to the subjective experiences any more strongly than the objective measurements, so I don't really see the point of it

I think that is because you are far too focused on the subjective experience. It is really trivial, scientifically-speaking. The scientifically interesting portion of such a model is its experimental predictions. The subjective experience would only be useful insofar as it simplifies the predictions or the derivation. Just like the one way speed of light in relativity.

 

[Dale]

@A.T. I don’t want to give the impression that I believe this is how a theory of the mind will work or will be developed. My point is just that the objections you have raised apply to validated theories of physics. The problems with a theory of the mind are the poor measurements and high complexity, not the unmeasurable aspects of experience. Every major theory has such unmeasurables.

 

[A.T.]

@A.T. I don’t want to give the impression that I believe this is how a theory of the mind will work or will be developed. My point is just that the objections you have raised apply to validated theories of physics.

And, I don't want to give the impression that I object to the approach, to care only about we what can objectively measure. That is perfectly fine and as science is usually done. I am just wondering if this approach will ever be able to answer questions about subjective experiences like "Do I subjectively perceive the color red the same way as you do?".

Of course you can postulate, that all normal sighted people subjectively perceive the color red the same way. But then using the correctness of predictions as evidence that the postulate is correct, is not that obvious to me:

1) The postulate must actually affect the predictions of the model. Otherwise you could take SR, add a random 3rd postulate, and claim all tests of SR support your random claim.

2) You can have multiple theories with wholly different postulates that arrive at the same predictions. So I don't think correct predictions prove that the postulates, which were useful for those predictions, have to be seen as objective truths.

 

[Dale]

I am just wondering if this approach will ever be able to answer questions about subjective experiences like "Do I subjectively perceive the color red the same way as you do?".

Such questions are philosophical rather than scientific. So the situation would be the same as with current philosophical questions addressed by scientific models.

The postulate must actually affect the predictions of the model.

Would you say that the one way speed of light “actually affects” the predictions of SR?

You can have multiple theories with wholly different postulates that arrive at the same predictions. So I don't think correct predictions prove that the postulates, which were useful for those predictions, have to be seen as objective truths

Agreed. But someone who does accept them as objective truths can never be proven wrong empirically.

 

[A.T.]

Such questions are philosophical rather than scientific.

I'm not sure they fit in philosophy either, but this is just a categorisation issue.

Would you say that the one way speed of light “actually affects” the predictions of SR?

Would the theory work without it?

 

[Dale]

Would the theory work without it?

The math would be far more complicated, but the experimental predictions would be the same.

So would you say that the one way speed of light “actually affects” the predictions of SR?

I'm not sure they fit in philosophy either, but this is just a categorisation issue.

Yes. I was just using “philosophical” for a catch-all term including anything that is not scientific. So it could be philosophical, aesthetic, religious, conventional, political, or anything else. I just lump it all in as “philosophical” out of laziness.

 

[Mayhem]

I have a little experience in molecular dynamics, which deals with how molecules (or atoms) behave as system, and may be formulated with classical physics. An introductory simulation deals with how phospholipids suspended in water spontaneously assemble into a sheet-like lipid bilayer, eventually settling at equilibrium.

In the simplest case, such a simulations typically involves thousands of individidual water molecule and hundreds of phospholipids. One could simulate the motion of each molecule individually, and then record the emergent behavior. However, this naive approach quickly becomes unfeasible as the number of molecules increases, and large simulations are instead made possible by simplifying the system anywhere it does not significantly sacrifice experimental relavance.

Now imagine we do not just want to see how phospholipids assemble into sheet-like bilayers, but also how these bilayers evolve and form something resembling a cell membrane. The scope of the problem goes from efficiently simulating clusters of molecules to predicting the behavior of the emergent structures of the system. In principle, both phenomena could be studied in the same simulation if we ensured that there are enough particles to allow for larger structures to form, and leave the simulation running long enough.

However, in the same way simulating individual atoms becomes unfeasible for simulating clusters of molecules, simulating macrostructures from individual clusters of molecules becomes unfeasible. Any simulation will therefore be optimized for accuracy in the domain of interest. Typically, accuracy at small scales sacrifices accuracy on large scales and vice versa.

The first point I want to make is how limited we are by computation in understanding biological systems on the scale of atoms and molecules. Consciousness is to our best understanding is an emergent phenomenom of the brain, which itself consists of billions of neurons, each of which consist of trillions of molecules (trillions!), in a very complicated and non-arbitrary network. It follows that any scientific understanding of consciousness starts, at best, is limiteed to the resolution of the neuron.

Next, consider how many brains exist that have no verifiable indicators consciousness, such as the brains of simple animals or even a third trimester human fetus (if not newborns, but I'm not opening that can of worms). These arguably unconscious brains are still well beyond our understanding. I think it is safe to say that science is simply not ready for the "hard problem of consciousness".

 

[A.T.]

The math would be far more complicated, but the experimental predictions would be the same.

So would you say that the one way speed of light “actually affects” the predictions of SR?

Based on the above, I would say no, it only affects the complexity of the calculations, not the predictions.

Yes. I was just using “philosophical” for a catch-all term including anything that is not scientific.

But even the boundary of "scientific" is not clear. Medical science deals with subjective experiences, including hallucinations.

A hypothetical complete scientific theory of the mind, combined with hypothetical complete measurements of the brain, could be able to tell what a person is dreaming or hallucinating, even tough these are completely subjective experiences.

The question is how far this description can go: It's one thing to correctly detect that a person is hallucinating pink elephants. But it's another thing to make another person experience the same pink that the hallucinating person is experiencing.

 

[Dale]

Based on the above, I would say no, it only affects the complexity of the calculations, not the predictions.

Then why would you insist that for a theory of the mind "The postulate must actually affect the predictions"? In physics theories, the postulates do not necessarily actually affect the predictions.

You are requiring more of a theory of the mind than is required of physics theories. That is what I am objecting to: expecting more of a theory of the mind than we expect of physics theories.

A hypothetical complete scientific theory of the mind, combined with hypothetical complete measurements of the brain, could be able to tell what a person is dreaming or hallucinating, even tough these are completely subjective experiences.

I think that you are changing what you have been meaning by "subjective experiences" here. I previously understood your intention for the term to include only specifically the unmeasurable aspects of experience.

 

[.Scott]

I think this whole question is working on the wrong end of the problem. Rather than worrying about how thoughts and emotions are built from atoms, first we need to fully model thoughts and emotions themselves. Without a comprehensive mathematical model of thoughts and emotions the scientific problem is undefined. We understand atoms far better than we understand thoughts and emotions. So that is the end of the problem to work on.

I will go one up on this.
A possible, perhaps likely, way of understanding human consciousness will be to: 1) model human thoughts, emotions, decision-making, etc. on a computer; 2) implement the design in hardware, including quantum processing components; 3) compare it to human responses and remodel and reimplement as needed; and finally 4) use that model and design to assist in determining exactly how our brains operate.

Oh, and don't implement anything that would worry about being turned off or recycled - it doesn't have to be that perfect of a human model.

 

[A.T.]

Then why would you insist that for a theory of the mind "The postulate must actually affect the predictions"?

The context was:

...using the correctness of the predictions as evidence that the postulate is correct...

In physics theories, the postulates do not necessarily actually affect the predictions.

If a postulate doesn't affect the predictions, then you cannot you use the correctness of the predictions as evidence that the postulate is correct.

You are requiring more of a theory of the mind than is required of physics theories.

No, I think the above applies generally.

I think that you are changing what you have been meaning by "subjective experiences" here. I previously understood your intention for the term to include only specifically the unmeasurable aspects of experience.

I used "subjective experiences" as "not objectively measurable from the outside". And medical science is currently dealing with things that are currently not objectively measurable from the outside.

 

[Dale]

No, I think the above applies generally.

OK, that is fair enough. Your position is not the usual position, but it is self-consistent. I don't object to that, I just object to inconsistency.

I used "subjective experiences" as "not objectively measurable from the outside". And medical science is currently dealing with things that are currently not objectively measurable from the outside.

Then "what a person is dreaming or hallucinating" cannot be "subjective experiences". Dreams and hallucinations are measurable from the outside with surveys, interviews, journals, and physiological measurements.

 

[Motore]

The question is how far this description can go: It's one thing to correctly detect that a person is hallucinating pink elephants. But it's another thing to make another person experience the same pink that the hallucinating person is experiencing.

Well we are kind of doing this already, I'm sure you already heard about direct brain-to-brain interfaces. For example, they use EEG to read the visual processing of one person, and TMS to physically stimulate the visual cortex of another, forcing the receiver to experience the exact same flashes of light.

We also already use fMRI and AI to decode neural patterns and reconstruct exactly what a person is looking at or even dreaming about onto a screen.
See here.

If we had a high-resolution neural link to write that decoded pattern directly back into your visual cortex, you would experience the exact same pink elephant.