TL;DR
- Part 1: Explores how the brain leverages the electromagnetic field to create a real-time world simulation through non-linear wave computing.
- Part 2: Demonstrates the explanatory power of this theory using:
- Wave-like behavior in perceptual experience.
- Resolution of computational challenges, such as spatial auditory processing.
- Accurate implications regarding visual perceptions.
- Part 3: Speculates on the mechanics behind the dynamic construction of the simulation.
- Part 4: Connects these mechanics to the nature of the self, non-duality, and the future of consciousness research.
- Part 5: Discusses the epistemic status of the theory and potential future research directions.
Part 1 - The Simulation
The brain–the most celebrated human organ–allows for abstract reasoning, fine motor control, navigation of complex social environments, and, of course, learning. It is no wonder that natural selection has evolved this tool to best our fellow Darwinian competitors. Of all of these perks, one feature stands out as particularly evolutionarily useful: the ability to instantiate a real-time world simulation (Lehar 2010) as depicted in the image below.
This world simulation is complex, immersive, and allows us to reason spatially in real-time, an enormously useful ability for survival (how else would you be able to duck under branches and leap over fallen trees while running away from predators in the jungle?).
This simulation is so useful that evolution decided it was worth building a heavy, delicate, and calorically expensive machine (consuming around 20 watts! (Kováč 2010)) that requires charging 8 hours a day to run it.
One of the most enjoyable ways to grasp the “simulationness” of your experience firsthand is by listening to a symphony on a pair of high-quality speakers. Instead of perceiving the sound as originating solely from the speakers, the brain creates an illusion of spatial positioning, placing the instruments in different locations within your perceptual space. Good speakers quite literally hack your brain into projecting fraudulent spatial information onto the entire world simulation.
What is this simulation made out of though? What are its functional parts? How does a squishy, wet, and electrically intertwined sponge generate the 3D simulation you are inhabiting right now? The Brain as a Non-linear Optical Computer (BAANLOC) theory proposes that the brain enlists electromagnetic waves to create a powerful computational medium. In other words, evolution has co opted the EM field for its massive parallelism, holistic behavior, and self-organizing tendencies, to unleash its computational potential with non-linear wave interactions. In this article we describe supporting evidence for the Brain as a Non-linear Optical Computer theory and speculate on how it may be implemented by existing neuronal mechanics in the brain.
Part 2 - Evidence for BAANLOC
Psychedelic Visuals
The most obvious example of the wave-like behavior in one’s experience may be the stereotypical visual effects of psychedelics, such as drifting: The Visual Effects of Psychedelics - broken down and described.
The undulation, breathing, and the geometric nature of psychedelic visuals suggests waves traveling over closed surfaces. Kaleidoscopic visuals, for example, can be explained by wave propagations that wrap back onto themselves. Waves traveling with little impedance on spherical surfaces generate geometric patterns via repeated self-interference. Below is a simple demonstration that creates psychedelic visuals purely via this mechanic.
A Natural Solution to Spatial Audio using Wave Computing
A practical example of wave computation comes from how the mind performs auditory localization. Test this out for yourself by approaching a busy street and closing your eyes. Pay close attention to the position, direction, and speed of the cars as they approach and pass you. You may be surprised how vivid and immediate the experience of space is, solely from your sense of hearing. The problem of localizing sound sources is being continually solved by the brain. Acoustic cameras attempt to solve the same problem using classical (digital) computers (see Acoustic cameras can SEE sound for an example). For a digital computer, however, the problem is so computationally heavy that it is difficult to achieve in real-time. So how does the human brain, consuming a mere 20 Watts while also processing visual information, abstract thoughts, maintaining homeostasis, etc, perform such a computationally expensive task in real-time? Wave computing of course!
We speculate that the brain’s solution to this problem is implemented by two wave-sources on opposite ends of a medium. The waves they generate correspond to the auditory stimuli from each ear. The direction of a sound creates a delay between the signals reaching each ear and consequently a phase shift between the wave-sources (illustrated below).
An interference pattern in the medium is produced and in the steady-state, the position of the sound source can be inferred by the phase of the standing waves. Wave computing solves this problem almost too easily. Below is an illustration of the steady state of the medium with a moving sound source.
Testing Implications of Wave Computing
BAANLOC also predicts certain behaviors. For example, one consequence of wave computing is that waves with higher frequency can resolve finer details. Assuming waves can travel throughout the whole medium, this implies that introducing high frequency information into your experiential field in one place will increase the resolution in other locations of that experience. We created this illusion to test this implication.
With your gaze focused on the cross in the middle of the left image, move your attention (but not your eyes!) to the stars in the periphery. Now do the same with the right image. Notice how the stars in the periphery become sharper if the star in the middle is sharp. This effect is unintuitive but easily explained by wave computing: when high frequency information is introduced to the computational medium, edges are resolved higher fidelity. This is similar to how higher frequency light is required to resolve smaller features in photolithography. An example of this phenomenon is below:
The images above are from an EM simulation of a rabbit being bombarded with waves from all four sides. The wave sources in the left image are low frequency while on the right they are emitting at a much higher frequency. You can see that the right image of the rabbit appears to be much more well-defined than the left one, due to the superior resolving nature of the high frequency waves (FDTD Simulation Library).
Given the supporting evidence, let us assume that the BAANLOC theory is correct and your brain is an optical computer that creates your world simulation. How would the brain actually implement this optical simulator?
Part 3 - The Electrostatic Brain: the Mechanics of the EM World Simulation
We propose that objects in your world simulation are made of patches in the neuronal lattice with distinct electrostatic parameters. The interaction of light with matter is governed by the material’s electrostatic parameters permittivity and permeability. Light propagates undisturbed through a uniform medium but reflects and refracts when these properties vary spatially, which is the principle behind how lenses manipulate light. Objects in your visual field may correspond to patches of neuronal substrate whose properties differ from those of the surrounding substrate, creating electrostatic boundaries between object and the background. EM waves would reflect off of these boundaries and resonate within the patch. Additionally, the waves inside the patch would have a different phase shift between their electric and magnetic fields. The figure below illustrates this.
How does the neuronal lattice generate these patches? Let’s back up and talk about neurons. The brain can be thought of as a configurable electron soup. Neurons create electric potential gradients by segregating cations and anions, resulting in relatively static patterns in the electric field. They also generate dynamic patterns when they fire (illustrated in the image below). First by the cascade of ions that rush to equalize the gradient (waves in the electron field), and second by EM waves created by the movement of those charges. Whether a neuron is at rest, is currently firing, or has just fired will change the electrostatic properties of the neural substrate in and around the neuron. Speculatively, a neuron with open ion channels may exhibit a relatively high permittivity perpendicular to the cell walls, resulting in a surface that reflects electromagnetic waves. Conversely, a neuron with closed ion channels will be transparent to EM waves, as the arrested charges cannot perturb the wavefront.
In other words, the electrostatic composition of neural substrates drastically affects what patterns in the EM field can exist. Of course, these patterns do not need to be explicitly “rabbit-shaped” in the sense of 3D structures in the EM field or the neuronal lattice and the electron soups they configure. Patches with different permittivity and permeability would only need certain geometric isomorphisms to the relevant physical shape, in order for those elements to be represented within and bound together with other elements of a world simulation - with isomorphisms potentially established over temporal as well as spatial dimensions. Investigating the minimal properties of such isomorphisms and looking for them in the brain is an exciting area of empirical and computational research that would lead to direct testing of this aspect of the BAANLOC hypothesis.
The relative permittivity of a patch of material also has an effect on the spatial frequency of the waves transmitted through that material. The diagram below shows a wave beginning in a region of low permittivity entering into a region of high permittivity. You can see the wave slow down and spatial frequency increase as it moves into this region.
The spatial frequency (density) of waves may encode scale. Patches with high permittivity and thus high spatial density may appear larger in our experience than patches of low permittivity. While the size of the patches themselves at the implementation could be more Euclidean, which may explain the duality of size perception described at this timestamp of Steven Lehar’s video and hyperbolic geometry of experience: The Dimensions of Visual Experience (Lehar 2003).
These are just some examples of how the neuronal substrate of the brain can govern the propagation of EM waves to construct a world simulation. If EM theories of consciousness hold any water, we suspect these mechanisms are the primary drivers of the character of our conscious experience.
Part 4 - An Electrostatic Model of the Self and Enlightenment
Finally, the perception of a “self” that is distinct from your environment may be rooted in the electrostatic configuration of the neural substrate. The instantiation of the “self” in the world simulation may be a portion of the neural substrate that is electromagnetically denser than its surroundings: a patch of neurons with heightened electrostatic permittivity which encapsulates EM waves and increases their spatial frequency (like in the previous figure). The equalization of the permittivity of this patch of neurons with the surrounding neurons may be the undoing of the privileged self and the mechanism responsible for the realization of anatta (non-self). An enlightenment experience is a moment where this happens and you realize you are the entire world simulation, not just the avatar within it. The figure below illustrates this phenomenon.
This perspective bridges ancient Eastern philosophy with contemporary neuroscience suggesting that meditation, psychedelics, and other consciousness-altering practices are radically modulating the electrostatic parameters of the brain, altering the dynamics of wave propagation and diminishing the boundaries between “self” and “non-self”. This claim may currently or in the near future be testable with high precision neuronal probing and ultimately this kind of research may illuminate the true nature of the self and shed light on the state space of consciousness.
Part 5 - Epistemic status and where all this goes next
The purpose of this post is to explore how the brain might exploit electromagnetic fields to create a real-time world simulation via non-linear wave computing.
The idea is neat in potentially providing a unified explanation for several phenomena discussed above, including the unified nature of the world simulation, the relative locations of simulated elements, sound localization, drifting effects in psychedelic visuals, and wave effects in certain optical illusions. Of course, other candidate explanations exist for these phenomena and the jury remains out for which explanations apply in each case. This post provides intuition pumps for fields having a significant role to play in at least some of these brain phenomena, but does not aim to argue against other explanations or claim a shut-door case. It might also turn out that fields are recruited for different phenomena than those listed here - our goal is to find these phenomena, wherever they may be.
In some cases, fields might play a complementary or implementational role alongside some other system mechanic. For instance, predictive processing can provide an explanation of the Fakhri Effect at a computational level, e.g. seeing one sharp star clearly might cause the brain to “predict” that other nearby stars in peripheral vision are also sharp. However, this abstract, higher-order, unconsciously cognitive phenomenon could be reinforced by more primitive ephaptic effects in the human visual system or world simulation modules. Or the EM field might be manipulated as a result of the predictive processing conclusion in order to implement it phenomenally.
Let’s recap on where this idea fits with other scientific findings in brain science. It’s worth reminding ourselves of the default reason that mainstream computational neuroscience doesn’t (yet) pay much attention to field effects: standard undergrad teaching emphasizes that myelin sheaths on neurons effectively shield them from EM field effects, such that you can focus on the input dendrite signals and the activation rules for conducting a signal out through the axon along to outward terminal connections. Overall fields are also considered extremely weak compared to the electrical and chemical signals that are known to mediate neural functions. None of this denies that an external, artificial, targeted EM field could not be strong to activate a cluster of neurons and change conscious experience (e.g. TMS is well-evidenced) - but it does deny that the brain is endogenously producing and channeling such fields to deliberate effect as part of its ordinary functioning.
What is our response to this skepticism? To a first order approximation, it may be reasonable to model neuronal functions without EM field effects. We welcome those research directions and want more of it. But it remains a highly simplified model. Fields are undeniably present and a higher order approximation would need to include them, if only to demonstrate that the myelin is adequate to shield all possible effects. More importantly, there is more to the brain than myelinated axons - and plenty of other areas where fields can get a causal grip on activity. As a few examples, dendrites (receiving input signals), axon hillocks (related to action potential initiation), and axon terminals (outbound signals), as well as Ranvier nodes along the axonal pathway - these neuron components are typically not myelinated. Outer cortex layers, gray matter, some intracortical connections and connections within the hippocampus also feature less myelination. Myelination may be more about structural integrity and speed over long distance communication than excluding informational “disruption” via surrounding fields, leaving plenty of scope for meaningful informational roles across different brain functions. With the brain often best analyzed as a “critical system” (Tian et al. 2022), even small effects near tipping points can have major consequences.
Aggregate field effects measured from outside the brain may be quite weak or mixed, but that does not prevent there being causally relevant, tuned, and highly-differentiated field effects for individual cells or clusters of cells within the brain. From an evolutionary perspective, EM fields are produced as a “free” consequence of EM activity elsewhere in the brain. If it is possible to do something useful with them, it is plausible that some evolutionary routes would want to do so.
There is already growing usage of such higher quality models using field effects to generate valuable findings in neuroscience. (McFadden 2013) provides a discussion of early research of non-epiphenomenality in EM fields and (Hales 2014) discusses fields in the context of pyramid neuron models. In more recent work, (Pinotsis, Fridman, and Miller 2023) draw together work on ephaptic coupling: showing how electric fields sculpt neural activity in the context of brain infrastructure, potentially tuning it to process information more efficiently, as well as influencing memory formation (Pinotsis and Miller 2023). Collective neuronal behavior in the form of oscillations may also be coordinated in part through field effects, with such oscillations identified with a range of potentially useful functions (Hunt and Jones 2023).
Our ideas are speculative, but testable. As with most areas of consciousness research, the most accessible tests address some combination of the core hypothesis and auxiliary assumptions (Fazekas, Cleeremans, and Overgaard 2024). We’ll have more on this in a future post, but to provide one illustration for now: wave effects in the Fakhri Illusion. There are multiple possible causes of the Fakhri Illusion, but we can shift our credences towards primitive wave effects and away from predictive processing effects. The predictive processing effects rely on higher order brain functioning, which it may be possible to remove or slow down in some settings, e.g. via chemically-induced states or advanced meditative states of de-reification which might correspond to a breakdown of certain predictive processing mechanisms. If the illusion remains present, we should be more confident of a primitive effect. Brain lesions in patients or animal subjects might achieve a similar effect, provided we can still find a way of subjects reporting perceptual differences, perhaps in more dramatic versions of the illusion.
Alternatively, with an auxiliary assumption around the specific EM field structures involved, we may be able to predict differing strengths of variants of the Fakhri Illusion by altering the relative size of objects/edges, changing the distance to the periphery, or introducing additional objects. These differences would result in different EM field topology, patterns of permittivity, and wavelengths involved - with corresponding changes in phenomenological response to the illusion. By contrast, a minimal predictive processing model would predict no change in effect, since the core principle of “predicting that periphery objects are more like a central object” remains the same.
Even the core hypothesis is directly testable in principle - although it would need significant resourcing and some near future tech, along the same lines as described in (Gómez-Emilsson and Percy 2023). But absolutely doable in our lifetimes with the right commitment. We know what the world simulation is. You are almost certainly reading this post from within your world simulation, simultaneously aware of individual words, joint meanings, the device they are displayed on, and your broader environment. Levels of awareness of different elements can vary, especially when entering flow states, but they can be called upon when needed. We also know the world simulation can be disrupted and some meditative, clinical, or psychedelic states can have minimal world simulations - perhaps zero world simulation - alongside persisting consciousness.
The electromagnetic activity correlated to the world simulation can be identified experimentally, through brain monitoring, analysis of lesions, computational modeling, and medical interventions. That correlated activity will have multiple components: neuronal activations, neurotransmitter patterns, interconnections with other nervous system components, electromagnetic fields, and so on. To identify which are epiphenomenal exhaust and which are causal, we can turn some off (or at least modulate them) and ask what happens to the conscious experience. It is possible to maintain neuronal activation patterns while changing the electromagnetic field around it, e.g. through targeted external stimulation that remains below thresholds to activate new neurons. In one theory, your world simulation is unaffected - in ours, something would change. No-one said it would be easy, but in a world of supercolliders and reusable rockets (not to mention US$100+ billion of revenue in the film industry, it shouldn’t be declared impossible).
Acknowledgements
This work is heavily inspired by the work of Steven Lehar (Cartoon Epistemology and The Grand Illusion), Andrés Gómez Emilsson (Brain as a Non-Linear Optical Computer), and Johnjoe McFadden and Susan Pockett (Electromagnetic Theories of Consciousness).
Tags
Consciousness, Psychedelic, Electromagnetic Field, Neuroscience, Non-duality, Visual Perception, Wave Computing