Roger Penrose more than 10 years ago said this was how it worked. I have spoken/thought about this before I knew Penrose thought of it but Penrose put the meat to the bone describing the location of memory and showed how brain waves frequency could be computed from quantum theory.
It’s been obvious to me for some time that the brain is a quantum computer. For me the reasons are the following.
After 50+ years of trying to build learning machines we are basically still not even at the chicken level of intelligence or learning.
In spite of the hype about AI the actual state of the discipline is pretty meager. What we have done is improve the recognition capability enough that we can do simple things like read letters and words, parse speech a little better, pattern match faces. Great hype is made of the latest AI advances but they are only slightly better than past approaches.
I used this analogy in a paper I wrote as undergraduate thesis at MIT. A chicken survives in the real world. It’s brain is very small but it reproduces, finds other chickens to mate with, evades predators. These predators are also learning and trying to figure out how to get the chickens.
Life is a constant battle and learning and behavior is far more complex than we assume when we see these creatures. They know what foods to eat and to avoid. Their behavior adapts as they evolve but also during their lifetimes. Don’t fool yourself into thinking that we really understand or could build a “chicken.” We can’t. We are very very far from building a machine that exhibits a fraction of the behaviors of even the simplest creatures.
After 40-50 years we have not identified what brain EEG patterns represent, how the brain stores memories, how it pattern matches against memories, how it learns things other than a gross idea that somehow neuron firings seem to be potentiated at synaptic junctions.
Our understanding of brain function is still stone knives. We have gross ideas of electricity moving, what chemicals are involved. We have made some advances but how it works at the base level of storing memories for instance we don’t really know. We can’t point to the exact place. We have theories and ideas but in general we haven’t been able to identify how the brian works. We have only recently found that electrical stimulation of the brain in certain parts can make some paralyzed patients be able to feel their limbs again. We aren’t sure why.
Basic functions that I think are crucial like are memories stored as molecules that are phosphorolized? Are they stored as potentiation by some chemical level? How does the brain do planning? Why? How does the brain learn concepts and store them or how does it link memories and feelings and smells to other things?
It is clear that even with billions of neurons there is no clear way the brain could store the vast amount of sensory information or even the summarization conceptualization of all the information it gets in the neuron system that has been proposed. It just doesn’t seem quanitfiably possible.
We have no idea how memories are stored in the brain. I mean the actual memories or concepts or whatever. The model proposed of the synaptic potentiating is a crude way recognition could work but it doesn’t describe how the actual memories are stored, i.e the actual sensory inputs which must be matched. Over a lifetime of vast inputs millions of sensory inputs/second there is an unbelievably large storage capacity and recognition capability but we’ve never found the hard drive. How do you have a computer without the hard drive? How do you say you understand anything about a computer if you don’t know where the hard drive is?
We see through our eyes a vast amount of information. Is all that stored or is only a fraction and why? How is it stored? At what level of detail? It is more than conceptual. We can reconstruct in dreams images. Sometimes we find things we didn’t see originally in a memory. Each sense and each nerve in the body is sending signals. The number of sensory inputs is millions every second this data is stored or some stripped version of this. On top of this are our thoughts and the abstractions built up from this base data are stored too. It’s recursive. The senses, the abstractions, the related feelings, the remembered things as echoes are also stored, the thoughts we had and the discoveries. Also, it seems our unconscious is somehow processing things and creating abstractions, reinforcing this and that.
We have no idea and we really have no conceptual framework organized for how to think of all this information but it is large. We don’t know how much is retained and how much is lost. Could it all be in there and just waiting to be unlocked?
We are able to trigger memories we didn’t think we still had. Just yesterday somebody mentioned some names from back in my college days many decades ago. The mere mention of those names brought back details of experiences, faces, events, feelings. A vast array of things that triggered other remembering. What is the limit of this system?
There has been no obvious mechanism of pattern matching clearly needed that has never been located or described how it could work.
One author I read who studied the brain said it was just a pattern matching machine. I agree 90% of what the brain does seems to be explained and understood as having to do this. However, nobody has been able to show how the brain could possibly do this.
The current theory of how this is done by potentiated neuron signaling chemicals doesn’t explain a lot. It explains one possible way that pattern matching could be done but it is a theory and it lacks proof and anything beyond the most conceptual high level description. We don’t know and haven’t observed if these potentiation are occurring the way we think. There is indirect evidence but it still leaves too much out. It doesn’t explain how abstractions get built and discarded. It doesn’t explain how linking this to memories works.
When we think it happens in leaps, i.e. we don’t think linearly. Our brains seem to operate more like the decoherence of a wavefunction and collapse to an idea rather than some logical process of a, b, c.
This is related to something Penrose has described. He says the brain does things that no logical machine could do. There are things no logic system can prove that humans can prove using a higher level conceptual framework. He is not even sure that a quantum computer could do these things.
I have noticed a similar problem. We don’t see during thinking that the brain works harder or is doing some massive search algorithm. It seems to happen almost magically ideas emerge into consciousness.
Creativity is the core function computers have not been able to emulate. Computers can iterate over all possible options but sometimes creativity involves going outside the box of known options. It requires linking other concepts and seeing connections between things that results in a conclusion, this is like that.
One of the big things computers and AI have failed at is seeing humor or satire. We want to be able to look through emails for instance and find the bomber. However, it is impossible for a computer to tell if someone is joking about blowing up this or that and so the computer just has no idea. Humor is one of the most creative things humans do and it is something computers are incredibly bad at doing or recognizing.
If Zuckerburg thinks AI will solve this problem and detect false or dangerous speech he is sorely mistaken.
Studies have shown that anesthesiology which suppresses consciousness works by suppressing microtubules in the brain
It is obvious to me that animals and even cells exhibit neural capabilities that are beyond our ability to explain.
Individual cells are able to learn and react to stimuli. If you look at animal behavior it is clearly creative and not automatonish. They are not stupid machines. They have feelings and react in many cases intelligently. They can be trained.
We have discovered that plants take advantage of quantum effects at high temperatures to improve the efficiency of capturing photons from the sun to use to power photosynthesis.
Using molecules called clorophores they are able to create a superposition of the electron to transport single photons.
More recently a study showed that birds who migrate when subjected to a magnetic field too small to move a single iron molecule were completely unable to fly along their paths. When the magnetic field is turned off they can go again.
Quantum properties have been identified in mircrotubules of the brain this year
I have had dreams where the complexity of the entire dream is as if it had been planned far in advance requiring active planning in advance of the events in the dream even though they seem to happen spontaneously in the dream.
All of these things have led up to this year a conclusion that the operation of the brain as a quantum computing machine to my mind has been proved.
This is amazing because the brain consists of billions of neurons, operating at room temperature and sometimes well above in a wet environment theoretically very difficult for quantum effects to be sustained for any period of time but it now appears clear that evolution has been able to use quantum computing all along from the plant times. It is therefore hard to understand how people would resist the obvious idea that nature has evolved to use molecules, atoms, DNA to fabricate quantum machines that leverage quantum effects.
The article is very detailed in how Penrose proposes this works although it is still missing key pieces but it moves the ball forward dramatically.
It is now clear that brain waves and the patterns we see are results of the decoherence and Penrose even is able to calculate some frequencies that match results we have observed.
This explains for instance the improvement in intelligence observed from meditation. Meditation or calming the brain seems to slow the beating of the brain and bring more coherence.
In Quantum physics what this means is that the neurons ( or microtubule components ) are able to stay in a state of superposition longer and thus able to make more powerful pattern matches and optimizations. This is logical. I have never understood why slowing the brain would increase intelligence. I could never understand how quieting or meditation could possibly affect brain performance. Now we understand that anything that can lengthen the time of superposition the more the quantum magic can happen. The more states that can be searched.
So many people have attacked Roger Penrose who in my opinion is one of the undisputed geniuses of our time. He has done things like prove the existence of Black Holes with Hawkins, contribute Spin logic to mathematics and other amazing mathematical discoveries.
Spin Logic is the basis for a new unification theory that creates quantization of space as well as the forces. This seems to me the most likely way that a unified grand unified theory will be created. He has contributed the mandelbrot to art and physics and mathematics and many other things. He has worked with Feynman and others on creating Penrose diagrams important for representing events in spacetime.
This guy, Penrose, is probably smarter than Einstein and he thinks its proved that the brain is a quantum computer. I think its proved as well. It has to be. However improbable something seems if it is the only way it could be done it must be the way it works. The brain was unexplainable before. This brings the possibility of understanding how the brain could possibly do what it does.
Consider that for $10M you can buy a DWave computer with 500 qubits. The human brain consists of billions of neurons each with who knows how many qubits but probably way way way more than 500. We are saying the human brain has the quantum computing capability at least a trillion times the speed of the DWave which recently was shown to be 5,000 times faster than networks of thousands of traditional computers at pattern recognition.
The brain is truly an awesome machine we can now see we are a long way from building a machine capable of being competitive with a human brain. That’s reassuring as a human that we aren’t likely to be replaced by machines anytime soon. I worried about that in the early days of AI. This discovery makes me feel like at least it will be a long time before we humans are disposable by the machines.