How do atoms build thoughts and emotions?

[frhaa]

Hi! Recently I´ve been wondering a lot about this. If atoms build everything and everyone, how are they capable of building human thoughts and feelings? It´s probably just a quick google away, but I thought I´d check this website out and hear what you guys have to say. Thanks!

 

[phyzguy]

Basically, no one knows. This is sometimes called "the hard problem of consciousness". It's fairly easy to understand how the neurons in your brain (which of course are made up of atoms!) can process images, respond to stimuli, etc. But why it gives rise to subjective experience like feelings is unknown.

 

[Paul Colby]

I was going to say one bond at a time but…

 

[Klystron]

Hi! Recently I´ve been wondering a lot about this. If atoms build everything and everyone, how are they capable of building human thoughts and feelings? It´s probably just a quick google away, but I thought I´d check this website out and hear what you guys have to say. Thanks!

Try taking smaller steps to understand human intelligence and emotions. Atoms combine to form molecules, molecules combine to form cells, cells combine to form multicellular (life)forms such as amoebas, et cetera. At some level of complexity 'atoms' lead to lifeforms including humans.

Learn basics of physics -- atoms and atomic structures, chemistry -- molecules and chemical bonds, and biology -- cells and multicellular structures leading to life as we understand it. One great leap involves understanding "What is life?". Study of biology leads to the question "What is human life? At what level of complexity does consciousness arise?". Study and understand what is known. Break the "hard problems" into digestible pieces that build understanding.

 

[lefty]

Hi! Recently I´ve been wondering a lot about this. If atoms build everything and everyone, how are they capable of building human thoughts and feelings? It´s probably just a quick google away, but I thought I´d check this website out and hear what you guys have to say. Thanks!

You would agree the brain is made of neurons, about 86 billion, all interconnected in a massively complex architecture? The Purkinge neurons in the cerebellar cortex receives about 100,000 connections from parallel neurons, and the substantia nigra dopamine neuron sends individual axions each exploding into massive arbors forming 2-3 million connections called synapses. Other neurons have 10-30 thousand synapses. And these neurons communicate with each other with chemical signals called neurotransmitters like dopamine, glutamate, serotonin and others which are relatively simple molecules made up of course with atoms (search the chemical structure of serotonin on line). When a neuron fires a nerve impluse through its axion, it secretes a neurotransmitter into the axion synapse or gap between axions of one neuron and dentrites of another neuron. These neurotransmitters chemically bind to the dendrite receptors and increase the likelihood of those neurons firing. The entire system is one massive chemical neural network.

Have you written one of the most simplest artificial neural networks on a computer where you use the mouse to draw a number in a square grid? You can train the network by drawing the number 2 with the mouse in many ways and the network will gradually learn what a 2 looks like by adjusting neural weights of the connections. Next, draw a 2 into the network you've never drawn before but still of course looks similar to a 2. The network, by virtue of the learned connections can identify the symbol you had drawn as a 2. It remembers. In the same way, the neural architecture of the human brain, by virtue of it's neural network connected by neurochemicals, through repeated neural firings, can adjust their connections and learns about the world in a similar way, a biological process called plasticity. Over time, these chemical adjustments encode memories, concepts, skills, and ultimately thought.

If you doubt this, consider the effect of LSD which has a chemical structure similar to serotonin. When someone ingests LSD, it enters the brain and "fools" the serotonin neurons into aberrant firings in the absence of what a normal biologically true event would cause the neurons to fire giving rise to an experience which does not exist in reality, i.e., hallucinations.

 

[berkeman]

(Mentor Note: Thread moved from the Atomic & Condensed Matter forum to the Biology forum for a better match.)

 

[ohwilleke]

You would agree the brain is made of neurons, about 86 billion, all interconnected in a massively complex architecture?

Lefty is on the right track. The path to wisdom here is thinking about the problem from the top down rather than from the bottom up.

(1) Start with the idea that thoughts and emotions are products primarily of the human body's nervous system.

(2) Then, consider the neurons and biochemicals that go into generating these thoughts and emotions.

(3) Then consider that the blueprints for neurons and the biological mechanisms are found in DNA and the biological processes by which DNA expresses itself in particular parts of the body to produce particular kinds of tissues and chemicals.

(4) Then consider the billions of years of evolution that caused the particular details of human DNA that makes this part of what the human body does come into being.

(5) Then consider how life arose from inanimate components.

(6) Then consider how the molecules that make up these inanimate components emerge from atoms interacting in particular conditions.

The question implies that going back to atoms is far enough and that one need not explain how atomic nuclei were formed (through a combination of Big Bang nucleosynthesis and subsequent nuclear reactions, mostly in stars that subsequently go supernova), or how protons and neutrons arose from quarks and gluons.

Once you break the question down into components, then you need to decide which of those six steps is really what you see as a mystery.

The hardest leap for a lot of people is step one. But in a world where we already create AI, and do genetic engineering, steps two or four may be a harder leap for some people.

 

[Demystifier]

The hard problem of consciousness arises from physicalism https://en.wikipedia.org/wiki/Physicalism , the view that physics should, in principle, be able to explain everything. If it can explain everything, then how can it explain consciousness? Without the assumption of physicalism, there is no demand that physics should explain consciousness, in which case the fact that nobody can't explain consciousness from physics is not considered a problem.

My opinion, as a physicist, is that physicalism is wrong. Physics is the most successful science, it explained really a lot, and it is precisely this success that seduced us to believe that it can explain everything. I've elaborated in more detail why I think that physics alone can't explain consciousness, here: https://www.researchgate.net/public...em_of_consciousness_Why_is_physics_not_enough

 

[Dr-Flay]

Atoms are the building blocks of elements, which is the physical domain.
Thoughts and emotions are woven from energy that resides in the meat-suit.

We have yet to discover how consciousness comes to be, so it is not really a question that can be answered at this time.
The 2 main camps of thought are that we either create consciousness and it is generated in us, or that it is made from a fundamental energy present in the Universe, and we act as attractors or harvesters.

The more I learn the more I am leaning into the camp of Sir Roger Penrose.

 

[renormalize]

The 2 main camps of thought are that we either create consciousness and it is generated in us, or that it is made from a fundamental energy present in the Universe, and we act as attractors or harvesters.

Can you cite one or more credible technical references that support the notion that consciousness "is made from a fundamental energy present in the Universe, and we act as attractors or harvesters"? Or is this rather a proposition from philosophy and/or religion?

 

[PeroK]

I have perhaps an unusual view that there is nothing special to explain. As a self-modifying algorithm becomes more sophisticated, it quickly become unpredictable and higher level properties emerge. These properties and concepts are needed to explain what the algorithm is doing. We get things like animal behaviour, which itself gets more and more sophisticated, until the behaviour of higher-level animals includes the ability to analyse themselves. This is basic consciousness. A bird might identify as part of a flock. Then, a more sophisticated bird, will be able to recognise itself as one of that flock. The only fundamental change is to take the step of turning your analysis not only on the rest of the universe, but on the part of the universe that is you.

Once animals progress beyond basic instincts, then consciousness, it seems to me, is almost inevitable as part of evolution.

 

[martinbn]

My opinion, as a physicist, is that physicalism is wrong.

Wouldn't this mean that there is some kind of magic?

 

[Demystifier]

Wouldn't this mean that there is some kind of magic?

It depends on what you mean by "magic". If by "magic" you mean something that can't be explained by the things we already know, then yes. If by magic you mean something that contradicts the things we already know, then no. The existence of consciousness does not contradict the laws of physics that we know. It only seems to contradict the assumption that those laws, by themselves, can explain everything.

 

[Dale]

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.

 

[Lord Jestocost]

The hard problem of consciousness arises from physicalism https://en.wikipedia.org/wiki/Physicalism , the view that physics should, in principle, be able to explain everything. If it can explain everything, then how can it explain consciousness? Without the assumption of physicalism, there is no demand that physics should explain consciousness, in which case the fact that nobody can't explain consciousness from physics is not considered a problem.

My opinion, as a physicist, is that physicalism is wrong. Physics is the most successful science, it explained really a lot, and it is precisely this success that seduced us to believe that it can explain everything. I've elaborated in more detail why I think that physics alone can't explain consciousness, here: https://www.researchgate.net/public...em_of_consciousness_Why_is_physics_not_enough

An excellent comment!

 

[PeroK]

It depends on what you mean by "magic". If by "magic" you mean something that can't be explained by the things we already know, then yes. If by magic you mean something that contradicts the things we already know, then no. The existence of consciousness does not contradict the laws of physics that we know. It only seems to contradict the assumption that those laws, by themselves, can explain everything.

Given how quickly AI is producing artificial versions of human behaviour, the hypothesis that consciousness is beyond what could be expected from the laws of physics is hard to justify. We may be very close to the scenario where a sufficiently complex AI system exhibits "artificial" consciousness".

Systems can talk about themselves. This is the basis of the proof of Goedel's incompleteness theorem, where he mapped statements about number theory to numbers.

I don't see anything mystical about consciousness. Only the next level of sophistication for an organism that has developed a complex, analytical brain.

 

[Demystifier]

Given how quickly AI is producing artificial versions of human behaviour, the hypothesis that consciousness is beyond what could be expected from the laws of physics is hard to justify. We may be very close to the scenario where a sufficiently complex AI system exhibits "artificial" consciousness".

Systems can talk about themselves. This is the basis of the proof of Goedel's incompleteness theorem, where he mapped statements about number theory to numbers.

I don't see anything mystical about consciousness. Only the next level of sophistication for an organism that has developed a complex, analytical brain.

Creating artificial consciousness is one thing, explaining it is another. Humans have been building houses before knowing the Newton laws of physics, and steam engines before knowing the laws of thermodynamics. This demonstrates that you can build something without a theoretical explanation of it. Likewise, if the near future technology would create a conscious AI, it would not automatically mean that we would have a theoretical explanation of it.

 

[PeroK]

Creating artificial consciousness is one thing, explaining it is another. Humans have been building houses before knowing the Newton laws of physics, and steam engines before knowing the laws of thermodynamics. This demonstrates that you can build something without a theoretical explanation of it. Likewise, if the near future technology would create a conscious AI, it would not automatically mean that we would have a theoretical explanation of it.

What's to explain?

If you ask someone how many people are in the room, including themselves, then I don't see the problem with them identifying themselves as one of those people.

That's the first aspect of consciousness.

Although, the real question is why any of this requires us to step outside the laws of physics, which imply the laws of chemistry, which imply the laws of biology, which imply animal behaviour and social structures and, eventually, individual consciousness. We don't need to know how the laws of physics imply animals that live in herds or hunt in packs in order to believe that the laws of physics can ultimately produce functioning animals. Intelligence is a further step - but why would we suspect intelligence and consciousness cannot be a product of the underlying laws of physics, chemistry, biology etc.

 

[Demystifier]

What's to explain?

If you ask someone how many people are in the room, including themselves, then I don't see the problem with them identifying themselves as one of those people.

That's the first aspect of consciousness.

I think robots can already do that, but that's not at all what is meant by phenomenal consciousness. At best, this could be some form of access consciousness, but the hard problem is to explain the phenomenal consciousness. https://protoscience.fandom.com/wiki/Phenomenal_and_Access_Conciousness

 

[PeroK]

I think robots can already do that, but that's not at all what is meant by phenomenal consciousness. At best, this could be some form of access consciousness, but the hard problem is to explain the phenomenal consciousness. https://protoscience.fandom.com/wiki/Phenomenal_and_Access_Conciousness

My answer is in the link you provided:

Many neurobiologists and computer scientists feel that philosophers such as Block and Searle are overly-pessimistic about the power of "computation", "program" or "algorithm" to produce human-like consciousness. The study of "computation", "program", "algorithm" and consciousness is too primitive for us to be able to trust our intuitions about exactly what is possible for computational algorithms to accomplish. Further, it may not matter what we call physical processes that can generate consciousness as long as we can figure out what they are and how to work with them. Thus, neurobiologists and computer scientists feel justified in continuing to search for the physical basis of consciousness and for ways to endow man-made devices with human-like consciousness.

I'm with them!

 

[gentzen]

What's to explain?

The basic challenge is to explain why Descartes‘ dualism is wrong. Just like the nonlocality of Newton‘ theory of gravity is wrong, as Newton himself was fully aware.
The harder challenge is to explain why the passive observer model of consciousness is also wrong. In Newton‘s time, we were passive observers of the solar system, which was explained so well by Newton‘s theory.

 

[A.T.]

Basically, no one knows. This is sometimes called "the hard problem of consciousness". It's fairly easy to understand how the neurons in your brain (which of course are made up of atoms!) can process images, respond to stimuli, etc. But why it gives rise to subjective experience like feelings is unknown.

I don't even know if anybody besides me actually perceives to be conscious, in the same sense I perceive to be conscious, or perceives to be conscious at all. How do you explain something objectively, that cannot be described objectively, as it is completely subjective.

 

[PeroK]

The basic challenge is to explain why Descartes‘ dualism is wrong. Just like the nonlocality of Newton‘ theory of gravity is wrong, as Newton himself was fully aware.
The harder challenge is to explain why the passive observer model of consciousness is also wrong. In Newton‘s time, we were passive observers of the solar system, which was explained so well by Newton‘s theory.

Proving a philosophy wrong is always a challenge. Descartes thought he could prove the existence of God. And, if someone wants to say that consciousness comes from God, then I can't disprove that.

All I can ask is where and how the laws of physics, fundamentally, cannot explain animals that have developed intelligence and consciousness?

God, mysticism, science. Take your pick, but don't knock science off the list!

 

[pinball1970]

I have perhaps an unusual view that there is nothing special to explain. As a self-modifying algorithm becomes more sophisticated, it quickly become unpredictable and higher level properties emerge. These properties and concepts are needed to explain what the algorithm is doing. We get things like animal behaviour, which itself gets more and more sophisticated, until the behaviour of higher-level animals includes the ability to analyse themselves. This is basic consciousness. A bird might identify as part of a flock. Then, a more sophisticated bird, will be able to recognise itself as one of that flock. The only fundamental change is to take the step of turning your analysis not only on the rest of the universe, but on the part of the universe that is you.

Once animals progress beyond basic instincts, then consciousness, it seems to me, is almost inevitable as part of evolution.

Me too.

Different brains, different types of CNS different levels of what we describe as conscious.

Damage or disease to the brain picks that apart in very specific ways over time, no more mysterious than liver disease preventing metabolism of vitamins, glucose, toxins etc.

 

[Dale]

All I can ask is where and how the laws of physics, fundamentally, cannot explain animals that have developed intelligence and consciousness?

Until we can experimentally measure those concepts (intelligence, consciousness, thoughts, and emotions) then indeed science fundamentally cannot explain them. All scientific explanations come through using the scientific method, the scientific method requires experiments, and experiments require measurements.

 

[Baluncore]

Consider a hierarchical tree. Emerging from each level are multiple properties.
We can climb that branching tree, until we emerge from the canopy, into enlightenment.
One such route, from atoms to consciousness, may go roughly as follows.

Molecular groups demonstrate the emergent property of atoms.
Self-replicating molecules are an emergent property of molecular groups.
Biological structures are an emergent property of self-replicating molecules.
Nerves and communication are an emergent property of biological structures.
Brains are an emergent property of nerves and communication.
Memory and internal feedback are an emergent property of brains.
Thinking is an emergent property of memory and internal feedback.
Consciousness is an emergent property of thinking.
Then we can climb further.
Reasoning is an emergent property of consciousness.
Philosophy is an emergent property of reasoning.
Mathematics is an emergent property of philosophy.
Physics is an emergent property of mathematics.
This forum is an emergent property of physics.

 

[PeroK]

Until we can experimentally measure those concepts (intelligence, consciousness, thoughts, and emotions) then indeed science fundamentally cannot explain them. All scientific explanations come through using the scientific method, the scientific method requires experiments, and experiments require measurements.

Behaviour, and levels of intelligence and consciousness can be measured in animals and humans. Perhaps not as perfectly as an electric current can be measured. Mental illnesses can be studied and, to some extent, affected by drugs or surgery. I fail to see the lack of science in this field.

As I said, you are free to choose the religious or mystic explanation for human evolution, intelligence and/or consciousness. I'm just asking that, especially on a science forum, science is not thrown off the list!

 

[Demystifier]

Until we can experimentally measure those concepts (intelligence, consciousness, thoughts, and emotions) then indeed science fundamentally cannot explain them. All scientific explanations come through using the scientific method, the scientific method requires experiments, and experiments require measurements.

To measure something means to use an empirical knowledge to associate a number (*) with it. In other words, science based on measurement is quantitative. By contrast, phenomenal consciousness is qualitative, it is associated with the notion of "qualia". That's precisely the source of the problem, how can a quantitative science explain something inherently qualitative?

Perhaps the problem of consciousness requires a totally new kind of science, based on qualitative empirical knowledge, rather than the quantitative one. Of course, we already have some sort of qualitative empirical knowledge, but it is rather vague and subjective. The problem is how to make it precise and objective so that we can make a serious science out of it.

(*) Curiously, the philosopher Hegel (1770-1831) defined the notion of measure as something associated with qualitative properties, rather than the quantitative ones. For instance, when the temperature of water changes from 99°C to 100°C to become a gas, he called it a change of measure. (The notion of phase transition in physics did not exist yet.) By the way, Hegel was an idealist, meaning that he thought that the essence of everything is the thought, rather than matter. He argued that numbers, formal logic and the notion of matter are just special types of mental ideas. In idealism, the existence of consciousness is not a problem. Instead, the problem is to explain the material physical world.

 

[Dale]

Behaviour, and levels of intelligence and consciousness can be measured in animals and humans

Behaviors and thoughts are different. Measuring one is different from measuring the other.

I fail to see the lack of science in this field.

So far the best instruments we have for measuring thoughts are surveys. They are really bad instruments from a metrological perspective. So it isn’t that there is no science here, but rather that our measuring instruments are really poor compared to the measuring instruments we have in other scientific fields.

As a result of our poor measurements we also don’t have a good ability to make and test scientific models. There are a variety of models of thoughts and emotions, but discriminating between them is experimentally difficult.

In my opinion this is what most limits us on these topics. Asking about how thoughts and emotions arise from atoms is putting the cart before the horse. We first need measurements of thoughts and emotions that are reliable, and validated models that explain and predict those measurements.

To measure something means to use an empirical knowledge to associate a number (*) with it. In other words, science based on measurement is quantitative.

Agreed.

By contrast, phenomenal consciousness is qualitative, it is associated with the notion of "qualia". That's precisely the source of the problem, how can a quantitative science explain something inherently qualitative?

This may be overstating it. We can associate quantitative numbers with these concepts. They are just really bad from a metrological perspective.

 

[Motore]

For me consciousness is an emergent phenomenon of higher and higher number of self-interaction neurons. There is no one neuron, one molecule that will be found responsible for consciousness. In that manner I actually see the brain more like an AI machine, where the brain has orders of magnitude higher number of self-interacting neurons. Here is a nice summary:
https://theness.com/neurologicablog/biological-vs-artificial-consciousness/