This is a fascinating exploration of how the metaphors we use to understand the brain can sometimes limit the questions we think to ask. The distinction between neural activity carrying information and actively participating in computation is particularly compelling. What I appreciate most is the careful separation between established findings, theoretical interpretation, and what still needs experimental validation. The possibility that brain waves are not merely background signals but essential components of cognitive coordination opens an exciting new direction for neuroscience. An exceptionally thought-provoking article that challenges conventional thinking without overstating the evidence.
One of the most intriguing aspects of this discussion is the distinction between reproducing intelligence and understanding consciousness. We often assume that if a system can reproduce the observable functions of the brain, it must somehow reproduce the underlying experience as well. Yet computation, biological mechanism, and subjective awareness may represent very different levels of explanation. The discussion of analog wave dynamics makes that distinction even more important, particularly as artificial intelligence continues to advance. I especially appreciate that the article doesn't rush toward definitive conclusions but instead identifies the scientific questions that could help distinguish competing explanations. This is precisely the kind of thoughtful, interdisciplinary analysis the conversation about AI and consciousness needs.
What makes this article especially valuable is its willingness to examine not only what a promising scientific theory might explain, but also where its limitations remain. The idea that the brain could use continuous electrical fields and traveling waves as active computational mechanisms challenges our tendency to interpret biological intelligence through the architecture of digital machines. Equally important is the emphasis on causation: observing a wave pattern alongside cognition is not the same as demonstrating that the pattern produces the computation. That scientific distinction deserves attention. I also find the proposed experimental tests particularly exciting because they give the theory opportunities to be challenged and refined. Thank you for presenting such a complex subject with intellectual rigor, clarity, and a refreshing sense of scientific humility.
The CPU power-draw analogy is the sharpest move in this piece. A wave pattern that tracks the computation closely enough to decode it still doesn’t prove the wave is doing the computing, and you’re right that intervention is the only way out. What I appreciate is that you let the theory stay a theory. The hypothetical experiment, disrupting interference patterns without silencing the neurons that generate them, makes the claim falsifiable, which is exactly where the analog layer either earns its place or collapses back into a measurement tool. Thank you for giving Buzsáki’s skepticism real weight rather than treating it as an obstacle.
Your four-part breakdown of what “reproduce” could mean (simulating a process, reproducing its functional organization, its physical mechanism, and its experience) is the clearest framing of that debate I’ve seen in a piece this short. I also liked that you declined to use the hurricane analogy as a knockdown argument, since whether consciousness behaves more like wetness or more like arithmetic is the very thing in dispute. The point that field dynamics turn “same computation?” from a thought experiment into an engineering spec is a genuinely useful sharpening. And the closing thought, that a system could pass every behavioral test while we still can’t say what it experiences, lingers.
The line about the bug that lives below the abstraction, balanced against the engineer who trusts no abstraction and never ships, captures the real tension in this field better than most academic framings. Neuroscience and AI both depend on deciding which omitted details matter for the question being asked, and the MIT proposal is essentially a claim that timing and field effects belong in the model. I appreciated the care in separating what is established (mixed selectivity, traveling waves, ephaptic effects in tissue) from what remains conjecture (the computational role of wave interference, and especially the leap to consciousness). That honesty about evidence levels is rare and makes the piece trustworthy. Excellent work.
Technology enthusiast exploring the intersection of AI, software development, business, and society. I write about emerging technologies, their real-world impact, and the opportunities and challenges
One of the most fascinating implications of this theory is that it challenges our assumption that computation is independent of the physical substrate performing it. We have become accustomed to thinking of computation as an abstract process that can be transferred between different hardware architectures without fundamentally changing its nature. But what if the brain's electromagnetic environment is not merely supporting computation but actively participating in it? That would introduce an important distinction between reproducing the mathematical behavior of a biological system and reproducing its actual computational dynamics. A digital simulation might perfectly model wave interference, ephaptic coupling, and neural synchronization without physically instantiating those interactions in the same way. This doesn't necessarily mean digital AI cannot be conscious, but it does mean we may be prematurely assuming that functional equivalence automatically implies equivalence of experience. Perhaps the question isn't simply whether machines can think like brains, but whether the physical process through which thinking occurs matters.
Technology enthusiast exploring the intersection of AI, software development, business, and society. I write about emerging technologies, their real-world impact, and the opportunities and challenges
I find the proposed relationship between brain waves and consciousness particularly interesting because it shifts attention away from individual neurons toward the dynamic relationships between them. Neuroscience has spent enormous effort identifying specialized regions, neural circuits, and information-processing mechanisms. Yet consciousness is not experienced as a collection of independent computations. It is experienced as a unified whole. The idea that traveling waves and phase relationships could dynamically coordinate otherwise distributed neural representations offers an intriguing potential explanation for this unity. It also raises a deeper question: perhaps consciousness depends less on how much information a system processes and more on how that information is integrated, synchronized, and continuously influenced across the system. If so, simply scaling neural networks or increasing their parameter counts may never be sufficient to reproduce the relevant properties of biological cognition. The critical ingredient might be the organization of interactions over time, rather than computational capacity alone.
Technology enthusiast exploring the intersection of AI, software development, business, and society. I write about emerging technologies, their real-world impact, and the opportunities and challenges
What strikes me most is that this theory could change how we think about the architecture of future AI systems, even if its claims about consciousness ultimately prove incorrect. Modern AI has largely followed the digital computational paradigm: represent information numerically, process it through layers, and optimize the system to produce desired outputs. Biological brains may instead exploit continuous physical dynamics, spatially distributed interactions, and multiple timescales of synchronization as intrinsic computational resources. This suggests an interesting research direction for neuromorphic computing and hybrid analog-digital architectures. Rather than merely building faster versions of conventional neural networks, we could investigate systems in which physical dynamics perform part of the computation themselves. However, I would be careful not to confuse biological inspiration with biological equivalence. Recreating brain-like waves would not establish consciousness any more than reproducing the electrical activity of a heart would establish that a machine is alive. The real breakthrough would be identifying which physical properties are computationally useful, which are necessary for cognition, and whether any are indispensable for subjective experience. That is a much more interesting engineering and scientific challenge than simply asking whether AI can pass as human.
The most valuable idea here isn't whether the theory is ultimately correct, but the distinction between a model that predicts behavior and a model that explains the mechanism.
It's entirely possible to have a useful abstraction that succeeds for decades and still discover later that it was hiding an important layer of the system. That's what makes these kinds of theories worth testing rather than dismissing outright. 👍
What I valued most is how carefully you kept the evidence tiers apart. Mixed selectivity is established, traveling waves are well documented, ephaptic effects are demonstrated in slices, and the computational role of wave interference is still a theory. That ladder is exactly what a topic this seductive needs. Your CPU power-draw analogy also captures the central difficulty better than most neuroscience write-ups do: a signal that tracks the computation so closely is not necessarily performing it. Buzsáki’s skepticism and the proposed intervention test, disrupting the interference pattern without silencing the neurons that generate it, show what a falsifiable version of this claim would look like. It is rare to see an article on consciousness say plainly that a marker is not a mechanism. Thank you for writing something this rigorous and readable.
The distinction between simulating a process, reproducing its functional organization, reproducing its physical mechanism, and reproducing the experience that accompanies it is the part I will be thinking about longest. The weather-simulation analogy is usually wielded as a knockdown argument, and you rightly note that it presupposes the very thing in dispute: whether consciousness is more like wetness or more like arithmetic. I also liked the framing that if field dynamics carry part of the computation, the question of whether a simulation counts as “the same computation” becomes less a thought experiment and more an engineering spec. Your closing image, a system that passes every behavioral test while we still cannot say what, if anything, it experiences, is a sobering place to land for anyone working on AI. A question for you: do you think the Butlin and Long indicator-property approach would need to change if wave dynamics turn out to be causally essential, or would it simply become another property to check for?
What I appreciate most here is the distinction between challenging an abstraction and abandoning computation altogether. “The brain isn’t running software” need not mean that cognition is beyond computational explanation. It could mean that the separation between program, processor, and communication medium is the wrong starting point. In the framework you describe, the activity of the system helps determine which interactions are possible next—the “execution” partly configures the machine doing the executing.
That makes the most interesting question, to me, not “analog or digital?” but “which causal relationships must an adequate model preserve?” If wave dynamics merely summarize what synaptic networks are already doing, they may offer a useful description without requiring a fundamentally different computational account. If they feed back into those networks in a task-specific way, however, leaving them out could mean omitting part of the mechanism rather than simplifying its implementation.
The anesthesia findings add an important wrinkle. The cited study did not find a simple disappearance of synchrony: both drugs increased low-frequency phase locking, while neighboring prefrontal regions became less phase-aligned and corresponding regions across hemispheres became more aligned. That makes “more synchronization means more consciousness” an inadequate summary of these results. The relevant feature may be the organization of coordination, not its overall amount. cell
I would also sharpen the proposed experimental test slightly: showing that an electric-field perturbation changes behavior would establish influence, but not necessarily the specific analog computation being proposed. A stronger test would derive a prediction from the proposed wave geometry—this phase shift should selectively disrupt this operation—and compare it against a competing account that does not assign a computational role to extracellular fields. Ideally, restoring the predicted pattern would rescue the operation. That would help distinguish a mechanism with explanatory power from another variable capable of disturbing a complicated system.
For AI, your distinction between simulation, functional organization, physical mechanism, and experience is the crucial safeguard. A digital simulation could undermine the claim that biological hardware is necessary for a capability without settling whether it is necessary for experience. Conversely, demonstrating that fields matter in brains would not, by itself, establish that artificial systems need the same fields.
The productive takeaway is neither “brains are just computers” nor “brains transcend computation.” It is that we should specify what our computational descriptions leave out, then ask whether those omissions change the system’s causal capacities. The possibility that the abstraction boundary itself needs revising is interesting enough without treating it as a solution to consciousness.
This is such an insightful and beautifully articulated piece. What resonates most is the idea that conviction and intellectual rigidity can look remarkably alike, even though one is rooted in principle and the other in self-protection. I especially appreciate the reminder that changing our minds should be seen as an expression of growth, not a confession of weakness. The real test of open-mindedness isn't how well we challenge other people's beliefs, but whether we're willing to examine our own when uncomfortable evidence appears. A powerful reflection on the humility required to keep learning. Outstanding work!
A wonderfully thought-provoking essay, and one that feels particularly relevant to the way we navigate disagreement today. The bridge analogy captures something important: when defending our positions becomes more important than solving the problem, even reasonable people can become obstacles to progress. I also love the distinction between remaining open-minded and remaining endlessly undecided. We need conviction to make difficult choices, but we also need the humility to recognize when new evidence calls for a different direction. Perhaps the greatest sign of wisdom isn't never being wrong, but creating the space—in ourselves and in society—to acknowledge it, learn, and move forward. Beautifully written and deeply worthwhile!
We're a place where coders share, stay up-to-date and grow their careers.
This is a fascinating exploration of how the metaphors we use to understand the brain can sometimes limit the questions we think to ask. The distinction between neural activity carrying information and actively participating in computation is particularly compelling. What I appreciate most is the careful separation between established findings, theoretical interpretation, and what still needs experimental validation. The possibility that brain waves are not merely background signals but essential components of cognitive coordination opens an exciting new direction for neuroscience. An exceptionally thought-provoking article that challenges conventional thinking without overstating the evidence.
One of the most intriguing aspects of this discussion is the distinction between reproducing intelligence and understanding consciousness. We often assume that if a system can reproduce the observable functions of the brain, it must somehow reproduce the underlying experience as well. Yet computation, biological mechanism, and subjective awareness may represent very different levels of explanation. The discussion of analog wave dynamics makes that distinction even more important, particularly as artificial intelligence continues to advance. I especially appreciate that the article doesn't rush toward definitive conclusions but instead identifies the scientific questions that could help distinguish competing explanations. This is precisely the kind of thoughtful, interdisciplinary analysis the conversation about AI and consciousness needs.
What makes this article especially valuable is its willingness to examine not only what a promising scientific theory might explain, but also where its limitations remain. The idea that the brain could use continuous electrical fields and traveling waves as active computational mechanisms challenges our tendency to interpret biological intelligence through the architecture of digital machines. Equally important is the emphasis on causation: observing a wave pattern alongside cognition is not the same as demonstrating that the pattern produces the computation. That scientific distinction deserves attention. I also find the proposed experimental tests particularly exciting because they give the theory opportunities to be challenged and refined. Thank you for presenting such a complex subject with intellectual rigor, clarity, and a refreshing sense of scientific humility.
The CPU power-draw analogy is the sharpest move in this piece. A wave pattern that tracks the computation closely enough to decode it still doesn’t prove the wave is doing the computing, and you’re right that intervention is the only way out. What I appreciate is that you let the theory stay a theory. The hypothetical experiment, disrupting interference patterns without silencing the neurons that generate them, makes the claim falsifiable, which is exactly where the analog layer either earns its place or collapses back into a measurement tool. Thank you for giving Buzsáki’s skepticism real weight rather than treating it as an obstacle.
Your four-part breakdown of what “reproduce” could mean (simulating a process, reproducing its functional organization, its physical mechanism, and its experience) is the clearest framing of that debate I’ve seen in a piece this short. I also liked that you declined to use the hurricane analogy as a knockdown argument, since whether consciousness behaves more like wetness or more like arithmetic is the very thing in dispute. The point that field dynamics turn “same computation?” from a thought experiment into an engineering spec is a genuinely useful sharpening. And the closing thought, that a system could pass every behavioral test while we still can’t say what it experiences, lingers.
The line about the bug that lives below the abstraction, balanced against the engineer who trusts no abstraction and never ships, captures the real tension in this field better than most academic framings. Neuroscience and AI both depend on deciding which omitted details matter for the question being asked, and the MIT proposal is essentially a claim that timing and field effects belong in the model. I appreciated the care in separating what is established (mixed selectivity, traveling waves, ephaptic effects in tissue) from what remains conjecture (the computational role of wave interference, and especially the leap to consciousness). That honesty about evidence levels is rare and makes the piece trustworthy. Excellent work.
One of the most fascinating implications of this theory is that it challenges our assumption that computation is independent of the physical substrate performing it. We have become accustomed to thinking of computation as an abstract process that can be transferred between different hardware architectures without fundamentally changing its nature. But what if the brain's electromagnetic environment is not merely supporting computation but actively participating in it? That would introduce an important distinction between reproducing the mathematical behavior of a biological system and reproducing its actual computational dynamics. A digital simulation might perfectly model wave interference, ephaptic coupling, and neural synchronization without physically instantiating those interactions in the same way. This doesn't necessarily mean digital AI cannot be conscious, but it does mean we may be prematurely assuming that functional equivalence automatically implies equivalence of experience. Perhaps the question isn't simply whether machines can think like brains, but whether the physical process through which thinking occurs matters.
I find the proposed relationship between brain waves and consciousness particularly interesting because it shifts attention away from individual neurons toward the dynamic relationships between them. Neuroscience has spent enormous effort identifying specialized regions, neural circuits, and information-processing mechanisms. Yet consciousness is not experienced as a collection of independent computations. It is experienced as a unified whole. The idea that traveling waves and phase relationships could dynamically coordinate otherwise distributed neural representations offers an intriguing potential explanation for this unity. It also raises a deeper question: perhaps consciousness depends less on how much information a system processes and more on how that information is integrated, synchronized, and continuously influenced across the system. If so, simply scaling neural networks or increasing their parameter counts may never be sufficient to reproduce the relevant properties of biological cognition. The critical ingredient might be the organization of interactions over time, rather than computational capacity alone.
What strikes me most is that this theory could change how we think about the architecture of future AI systems, even if its claims about consciousness ultimately prove incorrect. Modern AI has largely followed the digital computational paradigm: represent information numerically, process it through layers, and optimize the system to produce desired outputs. Biological brains may instead exploit continuous physical dynamics, spatially distributed interactions, and multiple timescales of synchronization as intrinsic computational resources. This suggests an interesting research direction for neuromorphic computing and hybrid analog-digital architectures. Rather than merely building faster versions of conventional neural networks, we could investigate systems in which physical dynamics perform part of the computation themselves. However, I would be careful not to confuse biological inspiration with biological equivalence. Recreating brain-like waves would not establish consciousness any more than reproducing the electrical activity of a heart would establish that a machine is alive. The real breakthrough would be identifying which physical properties are computationally useful, which are necessary for cognition, and whether any are indispensable for subjective experience. That is a much more interesting engineering and scientific challenge than simply asking whether AI can pass as human.
The most valuable idea here isn't whether the theory is ultimately correct, but the distinction between a model that predicts behavior and a model that explains the mechanism.
It's entirely possible to have a useful abstraction that succeeds for decades and still discover later that it was hiding an important layer of the system. That's what makes these kinds of theories worth testing rather than dismissing outright. 👍
What I valued most is how carefully you kept the evidence tiers apart. Mixed selectivity is established, traveling waves are well documented, ephaptic effects are demonstrated in slices, and the computational role of wave interference is still a theory. That ladder is exactly what a topic this seductive needs. Your CPU power-draw analogy also captures the central difficulty better than most neuroscience write-ups do: a signal that tracks the computation so closely is not necessarily performing it. Buzsáki’s skepticism and the proposed intervention test, disrupting the interference pattern without silencing the neurons that generate it, show what a falsifiable version of this claim would look like. It is rare to see an article on consciousness say plainly that a marker is not a mechanism. Thank you for writing something this rigorous and readable.
The distinction between simulating a process, reproducing its functional organization, reproducing its physical mechanism, and reproducing the experience that accompanies it is the part I will be thinking about longest. The weather-simulation analogy is usually wielded as a knockdown argument, and you rightly note that it presupposes the very thing in dispute: whether consciousness is more like wetness or more like arithmetic. I also liked the framing that if field dynamics carry part of the computation, the question of whether a simulation counts as “the same computation” becomes less a thought experiment and more an engineering spec. Your closing image, a system that passes every behavioral test while we still cannot say what, if anything, it experiences, is a sobering place to land for anyone working on AI. A question for you: do you think the Butlin and Long indicator-property approach would need to change if wave dynamics turn out to be causally essential, or would it simply become another property to check for?
What I appreciate most here is the distinction between challenging an abstraction and abandoning computation altogether. “The brain isn’t running software” need not mean that cognition is beyond computational explanation. It could mean that the separation between program, processor, and communication medium is the wrong starting point. In the framework you describe, the activity of the system helps determine which interactions are possible next—the “execution” partly configures the machine doing the executing.
That makes the most interesting question, to me, not “analog or digital?” but “which causal relationships must an adequate model preserve?” If wave dynamics merely summarize what synaptic networks are already doing, they may offer a useful description without requiring a fundamentally different computational account. If they feed back into those networks in a task-specific way, however, leaving them out could mean omitting part of the mechanism rather than simplifying its implementation.
The anesthesia findings add an important wrinkle. The cited study did not find a simple disappearance of synchrony: both drugs increased low-frequency phase locking, while neighboring prefrontal regions became less phase-aligned and corresponding regions across hemispheres became more aligned. That makes “more synchronization means more consciousness” an inadequate summary of these results. The relevant feature may be the organization of coordination, not its overall amount. cell
I would also sharpen the proposed experimental test slightly: showing that an electric-field perturbation changes behavior would establish influence, but not necessarily the specific analog computation being proposed. A stronger test would derive a prediction from the proposed wave geometry—this phase shift should selectively disrupt this operation—and compare it against a competing account that does not assign a computational role to extracellular fields. Ideally, restoring the predicted pattern would rescue the operation. That would help distinguish a mechanism with explanatory power from another variable capable of disturbing a complicated system.
For AI, your distinction between simulation, functional organization, physical mechanism, and experience is the crucial safeguard. A digital simulation could undermine the claim that biological hardware is necessary for a capability without settling whether it is necessary for experience. Conversely, demonstrating that fields matter in brains would not, by itself, establish that artificial systems need the same fields.
The productive takeaway is neither “brains are just computers” nor “brains transcend computation.” It is that we should specify what our computational descriptions leave out, then ask whether those omissions change the system’s causal capacities. The possibility that the abstraction boundary itself needs revising is interesting enough without treating it as a solution to consciousness.
This is such an insightful and beautifully articulated piece. What resonates most is the idea that conviction and intellectual rigidity can look remarkably alike, even though one is rooted in principle and the other in self-protection. I especially appreciate the reminder that changing our minds should be seen as an expression of growth, not a confession of weakness. The real test of open-mindedness isn't how well we challenge other people's beliefs, but whether we're willing to examine our own when uncomfortable evidence appears. A powerful reflection on the humility required to keep learning. Outstanding work!
A wonderfully thought-provoking essay, and one that feels particularly relevant to the way we navigate disagreement today. The bridge analogy captures something important: when defending our positions becomes more important than solving the problem, even reasonable people can become obstacles to progress. I also love the distinction between remaining open-minded and remaining endlessly undecided. We need conviction to make difficult choices, but we also need the humility to recognize when new evidence calls for a different direction. Perhaps the greatest sign of wisdom isn't never being wrong, but creating the space—in ourselves and in society—to acknowledge it, learn, and move forward. Beautifully written and deeply worthwhile!