Back in March 2021 I wrote about the Hard Problem of Consciousness which I consider to be the greatest mystery of all time. In that article I talked about what the hard problem of consciousness is - how can the physical brain produce a subjective experience, qualia? - and proposed a (speculative) mechanism by which the brain does it - the Chorus Hypothesis - just like in an orchestra or a chorus where the chorus is the combined result of the voices of the singers, subjective experience or qualia is the combined activity of the neurons which sync with each other and produce a pattern of waves - the brain waves. These waves do indeed vary by the state of consciousness (they are different in awake vs asleep individuals, for example). So there is at least that association, if not more. My hunch is that the subjective experience itself is the product of these brain waves (Christof Koch is also interested in this waving patterns in what he calls the local field potential). I think that the neurons do something special that the other cells in the body do not do (physically) and that's why they are the generators of consciousness - consciousness is the product of the physical activity of the neurons and my proposal is that the brain waves hold the key to producing it.
Now my initial article about the Hard Problem was already pretty speculative. I talked about the Chorus Hypothesis, why according to it only physical entities that produce brain-wave-like patterns can be conscious and how any kind of "simulated consciousness" is not conscious but a mere zombie - if you were to create a robot hardcoded to act like a human functionally it would not be conscious unless it would also have the physical activity that would create brain waves along with the functional behavior. A thermometer is not conscious, a computer is not conscious, the Internet is not conscious, a network of self driving cars is not conscious. Why? For one, they don't have brain waves. There is no "chorus" effect going on in these systems - just disparate information moving from one place to another. But let's assume I am mistaken with my chorus hypothesis - the brain waves are irrelevant - they are just mere artifacts that are completely irrelevant to consciousness. We then end up with the second point - the problem of information integration. In the aforementioned systems there is not integration of information whereas in the conscious experience it is - our consciousness includes all the senses at the same time - we see, hear, touch, smell and taste, all at once. They are not separate events - they are all integrated in a single experience. When I see something my hearing doesn't suddenly disappears because the conscious experience is busy hearing. I can taste an apple, listen to music and see my laptop screen while I complain in my head about this tooth pain I've been having these last few days. This entire information is integrated in a single, "instantaneous" subjective experience. Is this a problem? As we will see, it is.
My blog is named "aberatii intelectuale" - this effectively translates into "educated nonsense" from Romanian to English. In this article we will go into speculation squared territory and talk about a very lesser known problem of consciousness which is the spacelike separated neurons problem - this article will be about this problem in an "educated nonsense speculation" kind of way. Let me be clear from the start - this is not my original idea - I first became aware of this after Cristi Stoica's paper, "Are mental states non local?"
Weirdly enough, although the problem is pretty simple to understand, I have found that very few people actually listen to the problem and truly understand it. Usually people dismiss it out of hand, complain that it looks like a problem but it isn't (although they never come up with a solution) or just simply shrug their shoulders and think it's no big deal. I say it's a big deal. My goal in this article is to present the problem, to make it simple to understand by removing any complications and to present some scenarios of how the information flows in a human brain and why that is a problem in trying to come up with an integrated experience. Of course, you can reject from the get-go the entire idea of having our consciousness integrated but then I don't know what you mean - I certainly experience my inner life as integrated, just like I described above. Besides presenting the problem, I will also digress a little in some crazy ideas that I've been having (which are going to be speculation cubed, but we can have fun with that) regarding consciousness.
The Spacelike Separated Neurons Problem
OK, so let's attack the problem straight on. Let's think a bit about how the brain physically is constructed - here we don't necessarily care about the specific organization of the neurons and of neural networks, about the brain hemispheres, the corpus callossum that connects the two hemispheres, Brodmann areas or anything like that - not because these aren't very interesting things, but because we want to keep things simple - we can do this without undermining the spirit of the problem which is this: the neurons in the brain don't occupy the same location in space. They are not in a single point. They are distributed in different locations in the brain. OK, so? So what, right? Not so fast!
You see, due to the fact that the neurons are not in the same spot in space(time), as they communicate with each other it will take some time for each signal to reach from the source neuron to the target neuron(s). Why is that relevant? Well, each neuron is in its own reference frame (of course, the neurons are stationary to each other, in the brain). To put it more precisely, each neuron has its own causal light cone - events in spacetime (like other neurons emitting excitatory signals, inhibitory signals, emitting no signal at all etc) that are located in the past or future lightcones of these neurons.
So what does this all mean, exactly? Why should we care about this? Let's look at the following drawing (while immediately appreciating my artistic skills):
Relativistic lightcones
In the above illustration, the event E has a past lightcone and a future lightcone. These are the locations in spacetime that the event is being affected by from the past/can affect into the future. The cones themselves are constrained by the limited speed of light which is the fastest speed information can travel in spacetime. If we take a look at a few events, we can see a situation like this:
Connected and disconnected events in spacetime
Here we have three events: E1, E2 and E3. Events E1 and E2 are connected - E1 is in the past of E2 and E2 is in the future of E1. E3, on the other hand, being further away in space from both E1 and E2, is
not causally connected with either of these. In other words, because E3 is far away from E1 and E2, information about these events happening did not have time to reach E3, yet. Of course, as we allow time to evolve, the information from E1 and E2 will eventually reach E3 (and vice-versa) as the lightcones "grow". But it takes time for that to happen.
Picture this: "inflate" the brain to cosmic proportions (an example given by Cristi Stoica in his paper) - make the brain as big as a galaxy, with each neuron as a separate star. The neurons send information to each other as usual, it just takes a lot of time for the information to travel through the galaxy and reach each neuron - it takes thousands of years for neurons to exchange information. Now try to integrate the information in that system into a single experience - the subjective experience of an entity with a brain as big as a galaxy. This will make the problem a bit more obvious - how exactly do you integrate disparate information into a single experience? When a neuron emits information and changes state, no other neuron is aware of that change - the information hasn't had time to reach any other neuron, yet. As the information is traveling between the neurons (between the stars, in our example), does that entity have an actual subjective experience? What if, before the information sent by a neuron reaches the target neuron, the source neuron changes state? The target neuron will have obsolete information - the source neuron is not doing what the target neuron thinks it's doing anymore. Not only that, but how do the neurons know only to consider let's say these 3 neurons over there but not these other 5 neurons over there in constructing that subjective experience?
Let's make this more graphic - let's forget about the whole 86 billion neurons in the human brain and make the example as simple as possible: a brain made of just two neurons.
We have above one such example: there are only two neurons, Neuron A and Neuron B. A wants to send some information to B. Therefore, A sends a signal (say, an action potential) to B. This takes time to propagate - you can see that at T=0 the signal is still traveling between A and B and the same at T=1, whereas at T=2 the information finally reaches B. Great.
Now let's look at Scenario 2:
So what exactly is going on here? Well, A, just like in the first scenario, is trying to send some information to B. At T=0, A has sent the same information like in the previous T=0, the one in Scenario 1. Let's say that A is sending an active, excitatory signal towards B. But then at T=1, as the signal is traveling towards B, A sends a new signal - an inhibitory one. At T=1 you can see that both a blue and a red signal are traveling towards B (with the red signal being just emitted by A). None of the two signals have reached B.
At T=2, the blue signal reaches B. Before it reaches B, B sends a purple signal towards A.
So at T=2, the blue signal has reached B, the red signal is still traveling towards B and B has sent a purple signal towards A.
We can see in this simple example that from A's perspective, B has received a red signal - after all, that's the last signal that A sent. From B's perspective, B has received a blue signal - after all, B doesn't even know that a red signal has been sent - even at the speed of light, the red signal hasn't had time to get to B - so B does not know of the red signal's existence. Not only that, but from B's perspective, it has sent the purple signal so A (again, from its perspective) is purple. But from A's perspective nothing is happening - the purple signal didn't have enough time to reach it.
This might look confusing in writing but I think it's pretty clear in the illustration above. It's not a complicated system to understand and puts into an actual image the problem - there is no way to unify this situation into a single, unified, integrated experience. Whose perspective are you going to take in this whole configuration? A's? B's? Somewhere in the middle? No matter what your perspective is, the information is going to be contradictory. This is just a two neuron system - imagine what happens when you try to unify an 86 billion neuron configuration. Yet we say that "when the area X in the brain is active the subject experiences pain" or "the visual cortex appears active on an fMRI therefore the subject has a visual experience". We say that the brain activity is causing the conscious experience and we see from the story above that that cannot be so.
Now I already hear you bickering there in the background about my example. "What you're displaying here is just a static snapshot of the brain". "There's no subjective experience in a snapshot", I hear you say. "It's like taking a photo of a fire or of rain or trying to take a snapshot of a symphopy (or a chorus, since we're here) and complaining that there's a spacelike separated problem". Cool, let's take an interval of time, then. Hell, let's even take an example with waves or brainwaves - are we getting rid of the problem?
Take the profoundly beautiful illustration above. It's a view from above over our two neurons, A and B. This time we look at the waves the two neurons are doing, seen as concentric circles, here. The hashed red thing is where the two waves are interacting. Now let's consider Scenario 2:
In this less beautiful illustration Neuron A is changing the wave it's been emitting so far: for three cycles it's now been emitting the
blue wave, as the black wave is still propagating outwards to B (and everywhere else). Again, B has no idea that A has changed its tune - from A's perspective the wave currently emitted is the
blue wave. For B, however, the black wave is going to hit it first - B only knows about the black wave. We have the same problem - A has one state, B knows about a previous state. The system itself cannot be unified or integrated into a single experience.
Now you might think that maybe the wave itself matters - A and B are "stupid", they just produce physical effects - excite, inhibit, send action potentials etc. But it's the wave that matters. At first glance, this won't work. After all, if you sit into A's frame of reference, the wave looks in a way. If you sit in B's frame of reference, the wave looks different - B doesn't know how the wave looks at A's position.
I still don't know if this can be solved by looking at where the waves interact - maybe it's the interaction that matters - where the waves intersect. Maybe qualia and the hard problem of consciousness is solved by what the wave itself is doing - its frequency, maybe its amplitude, you know, wavy things. But then again I worry that you need to take a frame of reference from which to "look at the wave" - whose frame of reference is that going to be?
Yes, you are complaining again. First, you complained about taking snapshots and saying that conscious experience is a process, not a state. Now that we have the same problem for a process as well, even for a wave, maybe we're just overcomplicating things? Let's take rain - it rains everywhere, that's what raining means - the raining process is water drops falling from the clouds in a certain area. They are spacelike separated, yet rain doesn't suddely stop making sense, right? What about a computer? It has spacelike separated transistors, spacelike separated pixels on the screen and so on - yet the computer doesn't suddenly stop working just because of this, right? Despite having information spread out in certain portions of computer memory, your software still works, still displays the correct information on screen and you have access to an integrated information in your software. But do we?
When you look at the screen you do indeed see a unified screen. But that is only because it's your own consciousness that is doing the integration! The pixels on the screen are separated - it's a matrix of colors, each in its own frame of reference. There is no integration on a computer screen! Your conscious experience is doing the integration - it perceives these spacially separated pixels as "one single screen" - but they are not. There is nothing to explain when it comes to the computer screen - there is no "experience" associated with that process - it's your consciousness that is doing all the work. Same for rain - yes, it's a process, but each drop is separate from any other drop - it's your conscious experience that integrates the drops into one single entity. The same situation is also with a flock of birds that are flying in unison - the flock of birds looks like it's having a mind of its own - yet it's each bird masterfully navigating and avoiding collisions with all the other birds. Again, nothing to explain there - there is no subjective experience associated with that flock of birds - it's your conscious experience that "sees" the flock of birds as a separate entity - your conscious experience and the fact that it's integrated (or at least looks integrated) - that's what needs to be explained!
It gets worse: due to the relativity of simulatenity, the order of events differs from one observer to another. Therefore, some observer Alice traveling one way relative to the observed brain sees some different order of events vs observer Bob traveling the other way relative to the observed brain (this is true only for spacelike events - imagine neuron A sends signals to neurons B and C, B being on the left of A and C being on the right of A - if the order ABC or ACB matters in the subjective experience, different observers will either see ABC or ACB). Yet there will be only one subjective experience - the one which the subject itself is experiencing!
Other crazy ideas I've been having
Now that I have exposed what the problem is, time to get into speculation cubed territory and expose some crazy things that I've been thinking about - this won't be elaborate, just an enumeration of ideas, roughly.
For example, I've been wondering for some time if consciousness isn't actually a process of metabolism, not of "brain activity". This is probably not so, but maybe some metabolic process is also contributing - maybe the signals themselves are just mediating the metabolism in the brain - which is also registered on an fMRI. Maybe it's the metabolism that it's the trick (or part of the trick).
Another thing is this: it seems like all our sense are being propagated using the same exact mechanism. There are about 1000 different types of neurons, but the mechanism is the same - the signal is sent the same way and doesn't seem to carry with it any additional information - just neurons firing action potentials and exciting/inhibiting other neurons. There's one additional detail that needs to be taken into account - it seems the silence is also relevant. A neuron not saying something is significant - silence is important. It's not just something to be ignored - like a dog that didn't bark, that's an important detail - the dog didn't bark because the dog knew the guy. It's not an accident.
This silence is important detail seems very significant - somehow, the process of consciousness keeps track of this silence/activity thing going on in the brain. It's not just looking at activity but at silence as well. Again, it looks like the "chorus" of the brain is continuous, even when it's silent. Makes me wonder if we're truly at zero consciousness during coma, anesthesia or even death - we might have an experience and simply not remember it when we wake up - probably not, but still interesting to consider.
Then I was thinking about something else - if the neurons really don't actually propagate any information in the neural signal itself (maybe the signal's purpose is just to mediate metabolism or mediate the process of brain-wave generation), then how do different neurons produce different experiences? How can the neurons in the temporal lobe produce the experience of hearing while the neurons in the occipital lobe produce the experience of sight? What exactly is special about them, if the information of sight and hearing is not encoded into the signals from the sensory receptors in any way? And my crazy idea was that maybe it's the neuron's location in the brain that is important - what if, borrowing an idea from general relativity, the mind is the geometry of the brain? What if the shape of the brain itself and how the neurons are linked together geometrically is what gives rise to qualia and to the neural function of each brain area? That way, the neurons themselves can send bits of 1s and 0s (roughly speaking) - the information itself doesn't need to be differentiated. What is different is the location in the structure of the brain - the note that it produces in the chorus, like a musical instrument in a symphony.
Here's another one: maybe consciousness is possible as an analog process (as a wave) because the universe itself is a wave. We know that the classical world is artificial - it's most likely a product of entanglement - it looks classical because we are entangled with the observed items (particles etc). But in reality the universe is a wave. Maybe consciousness itself is a part of the wave function of the universe - it's a part of the wave function of the universe, the universe itself being a combination of these waves. Yes, I hear you, crazy idea.
OK, final one - how do you actually solve the problem of spacial separation of the neurons? Well, you could say that the conscious experience is not integrated, after all. Another option is to use the brain waves (maybe the intersection of them or something like that) as consciousness itself (why that is so is an ontological question that cannot be answered). Another way would be that consciousness is non-local - maybe the neurons are described by the same wave function and therefore are entangled and instantly "know" what the other neurons are doing - "know" in the sense of the entire structure is described by the same wave function - if you know A you also know B.
Do I know what the final answer is? Of course not. But it's fun to go through all these details and see if anything cool is not currently being considered by serious scientists. I think there is.
Comentarii