Wednesday, 10 February 2010

Easy read poster of the theory

>>>> AUTHORS NOTE: This section is still a bit of a mess .<<<<<


This website summarizes the progress of controversial "Dipole Neurology" theory (summarized here) which has exceeded my original expectations by developing from a spare time academic project (in neuropsychology) and has now moved into the search for a common physical pattern/topology for intelligence/consciousness (starting from the mammalian brain).  It is also developmental biology with accruing evidence backing up the basic concept. If basic becomes robustly proven,  I expect  this to become a strong contender in the top 10 of integrative neuroscience theories. 

The idea is controversial more by its scope and freshness, rather than proposing any quantum mind theory. The concept is proposing that the brain has a simple overarching structural framework (entire corticolimbic system) which is itself of a simple physical/computational form. This structural form is not the hodge podge of evolution neuroscience currently thinks, and is coherent enough to actually drive how the small scale computational elements (neurons/axons) etc end up filling up the large scale form for the overall highest level computation. The Cortical Dipole structure itself is proposed to enable feature extraction, as this works by brute force elements (neurons) operating through dipole laterality in a conceptually simple but well connected topology.  This structure arises in neurodevelopment to assist in guidance mechanics and is partly MHD (magnetohydrodynamic) in origin. From the neurodevelopment angle this concept is improving with new evidence from labs out there every year.

I have to remain open minded, but so far there does not yet appear as if there is any "real time" EM (electromagnetic) role in actual adult processing. At least nothing that could threaten current models. So for computation this project develops with traditional neuron/structure approaches, and then considers as an adjunct any EM aspects mainstream (not fringe) neuroscience has found. For example it is already accepted in science that magnetic structures have a variety of computational properties so from that we can consider what could a magnetic field contribute to information processing when it assists in and is integrated into the development of complex computational structures. 

This physics is important, because if computation is simple EM/thermodynamic both within and throughout the entire brains physical encapsulation then if true, this is a massive theory for neuroscience. But as i repeatedly will state, I don't actually explore the EM side too much in this project. The real controversy of this work is that it challenges the current idea that no topdown structural computational properties should be derived from physics and clearly assigned in general to the cortex or limbic system at the actual full structural level.

If that were not enough, this concept produces an inescapable finding or presumption (to be falsified). This is that if intelligence/consciousness comes from simple physics and gives rise to this specific type of dual process thermodynamic/EM physical structure I propose the brain has, then by logical conclusion this structure might be roughly applicable to the successful operation of most general learning systems where our laws of physics apply. So If this project continues as it has, it is moving to being able to make future testable predictions for the viability of Artificial General Intelligence (AGi) systems. That is whether AGi will have to possess something like a brainlike physical form as their complexity increases. Primarily due to constraints by the laws of physics.

Many science futurists predict that from 30 years from now, most primary technologies and later down the line, the core of most life itself will be reliant on complete AGi and mind understanding. My approach also provides a pioneers view of the brain which gives many original ideas. So its pretty exciting to be able to explore and lead these pioneering areas at such an early stage. Its also for this reason I had to become independent, but I kind of enjoy being a maverick now anyway :) I guess the only big question for newcomers here, is whether the cortex has dipole structure. Well first ask "what else could it be ?"

Check this beginners video for a quick visual intro, or open the poster further down

For this project I usually work alone as a new insight (or self criticism) comes to mind. As I am a student trying to develop this very complex work to the level of clarification required for laboratory testing, PhD and other applications I co-author peer reviewed publications with academics senior to myself. I also expose myself to group discussions and when affordable will present at a major neuroscience conferences with the aim of soliciting high quality criticism.

To explain the concepts I often generate simplified physical models for the cortico-limbic system as this appears to be the brains “information engine”, but do not take this to mean that other complex facts are ignored. Look at the cortico-limbic models here as the final overall "computational superstructures" in the same way we look at cortical columns as mini-computers. My most recent co-authored published application of physics to this simplified approach (see 2014 paper in links) tested whether fundamental thermodynamics is consistent with computational principles from entire structural morphologies.

The theories history summarized on this website will show a sequence of specific predictions has progressively lead from development of hypothesis to actual current theoretical framework. My proposal is that to develop and innovate a well worked physical topology of intelligence will be pivotal. Complex self developing AGi will eventually have to self organize into an optimal physical topology and I propose it will not be too different from the simplified form presented here.

See this blog post for a brief introduction - my 2013 paper for the first draft formalization of computational principles and my 2009 paper for the background neuro model. The next work planned will derive specific key facets, such as re-enforcement learning, spatiotemporal perception and self-awareness from the current simplified thermodynamic basis.


Although the brain structures are shown to be EM in nature, and the proposal here is the intelligence has a definite physical form it might be thought this is the grandad of all EM mind theories. But paradoxially this is not a hameroff style quantum mind theory. Or even like other EM mind theories. Although i have been open minded to some types of magnetic brain model for various interesting concepts and developmental mechanisms, so I do mention various respected authors on this site. But overall, I stick with traditional approaches and do not get into quantum mind theories, partly because i dont think they are as interesting as stuart hameroff et al., seem to think. The concept of quantum mind is over-rated, and appear to tempt people into various non-science thinking styles that are way off track from computation and more about imagination, desires to remain in obscure thinking or understandable drives to lead others through spirituality than solving problems or understanding neuroscience.

By contrast taking a "so what" attitude to QM mind, and concentrating on deriving a more boring thermodynamic formalization of this approach can steer us from those thinking pitfalls. This leads me to the conclusion we can have QM "style" processes in classical scales which we already perform on current digital systems as matrix problems. Also from the physics side quantum computing, there are recent findings of QM entanglement in classical fluids that show us fluid dynamics can solve quantum weirdness in terms of pilot waves. So it appears like wave/particle can be no big mystery. i.e. no quantum/classical divide after all. It seems getting into quantum mind is way off track if we dont need quantum computers to do what the brain does anyway.

This poster here was designed for hitting you with the theory in a get it quick visual manner. If you wish data, and more facets of this complex theory in depth there are the papers. And please not this theory is not derived from images and never was. Although even if it was that should not be a problem For this poster, Click here for full size (then click again). Cortex in centre is mirrored on coronal plane of the human brain to clarify the dipole structure.  The text underneath this image helps clarify the posters points.

In this poster I hope to illustrate that the strongest model for all of these cortex features to occur together in on developmental timeframe is a magnetohydrodynamic field model, which would assist in axon guidance and neuron migration for these larger brain structures.  If you read the  newer posts on this site you will find evidence is accumulating independently to verify that the astroglia which produce cortex structure have a magnetic field mechanism.

In response to peer review, it should be stated the cortex  dipole has all the features particular to magnetic or ferro-electric dipoles.  All these features cannot result from entirely electric dipoles (according to physics known so far). Also bear in mind all EM structure similarity is not actual live EM. It is what occurs when EM forces in neurodevelopment are  controlled, frozen and encapsulated within conventional biological details.

*    The axons have bent into a very clear magnetic dipole configuration; this would require a magnetic field that can pass through the biological tissue without interference. Corpus callosum and other axons in development are unmyelinated and the remaining axon components are ferroelectric  i.e. microtubules and voltage gated channels (refer to H. Richard Leuchtag). This means it is possible to propose axons are able to be influenced to move  in any magnetic dipole or linear MHD field present in the radial glia.

*     There is a broad domain wall at the longitudinal fissure. These domain walls are only known so far to result from magnetic or ferro-electric forces. i.e. Magnetic forces. I remain open minded if anybody finds domain walls dominated by the electric dipole, however as far we know virtual photon forces due to bound magnetic fields are the mechanism for domain walls.

*     The cortex gyri and sulci as well as the cerebellum (not in image), have appearance of variations of magnetic domain patterns (i.e. magnetic stripe domains). These patterns have correct scaling in accordance with neuron to glia ratios in various species (right of diagram). Variances in astroglia correlate to Variance in magnetic field strength (see mechanisms below). 

*     The cortical columns themselves have magnetic pinwheel type structures in the entire morphology at the boundaries and center. Similar pinwheels have been found in the ventricular zone radial glia by other researchers, so we have a facility to propose the internal formation of cortex columns by a MHD mechanism which is consistent from bottom to top layer of their developmental timeline.

*      Axons have an MHD soliton structure, and we know now axons are EM solitons. Neurons are not so easily defined in EM morphology terms (except looking like electrostatic branches) but they do follow the "greedy growth" principle which can be derived as least action which occurs at magnetic poles

*      To summarize in case this is not clear, every single part of large scale cortical morphology has a magnetic type feature, with a strong case for the mechanics in progress.

However it should be pointed out that traditional theories of brain structure and development exist  without an EM framework. i.e. Tension based,  because it is possible for that to occur. If we are being completely objective and entertaining what I say it is possible to say that any resemblance to magnetic structures are a collection of interesting co-incidences.  However developmental frameworks are currently incomplete and new works point to dipole forces being present. 

Dipole neurology also has a specific mechanism defined to make sense of confusion regarding chemoattractants and adhesion systems in development (see menu bar - neurodevelopment). Dipole neurology also provides computational frameworks for general computing which are also on track to make sense of our inability to define feature extraction in terms of neural coding. So with one physical approach neurodevelopment and computation become unified. So, If this theory is correct its a very big theory for neuroscience.

MAGNETIC MECHANISM:  There is recent acceleration of growing work for large neuronal activity magnetic fields to be produced by the mechanism of the astroglia ion flow i.e. Neuronal  Activity Associated Magnetic Fields (NAAMF) proposed by Marcos Banachlocha in 2001. More recently several others researchers are moving into this field. Lester Ingber a former Fenyman student who provides the statistical mathematics with Paul Nunez (a founding figure in EEG)  have produced papers in this regard.  Other Recent examples are  Bokkon & Banaclocha, 2010 , Pereira & Furlan, 2010, Størmer & Laane, 2009, 2011. I Estimate there may be about 25 papers since 2001 on this area with insistence for labwork grants on the most recent papers. However the problem is even though the biophysics and maths have now been  well produced by these researchers, it is very technically difficult to do such research on adult brains (due to magnetic interference) without building a specialized lab. (summary of these papers here)

ARRIVAL OF EVIDENCE IN NEURODEVELOPMENT : However for the purpose of neurodevelopment the news is more positive. The radial glia calcium ion flow proposed to give rise to the brain structures is actually just using the above NAAMF mechanisms, as the astroglia in adults are the remnants of the radial glia in neurodevelopment.  A study on development (Weissman, TA et al; 2004.) has found the calcium ions can pulse through the entire hemisphere radial glia in development. This would mean the NAAMF magnetic mechanisms can in theory apply to the entire cortex. (see more on that here). More recently last year the developmental biophysicist Vincent Fluery independently films a hydrodynamic dipole and quadrupole force dominating the entire development of the chick embryos skull.  This hypothesis is not then without a strong starting basis for evidence on several levels.  I cannot reveal details without permission but there is labwork planned to measure the brains field strength, pulses in times with the observed dipole /quadrupole hydrodynamic flow, so linking physical observation with underlying physics. 

From V. Fleury, 2011. A change in boundary conditions induces a discontinuity of tissue flow in chicken embryos and the formation of the cephalic fold Eur. Phys. J. E (2011) 34: 73

And more papers I was not aware off. Apparently there about a dozen or more papers studying the presence of an EM Field in neurodevelopment. Many of these papers propose that the axons are evolved to conform to an EM field.  More of that when this site is updated. For now read the references in this paper.

Text from above paper for the image below

"There are numerous hypothesized mechanisms by which cells might sense an electrical field. A weak electrical field could impose a force on negatively charged cell surface receptors, or alternatively the electric force imposed on positive ions (Na+) could result in the flow of their associated hydration shell, which exerts a drag force on cell surface membranes. The resulting asymmetrical redistribution of cell surface receptors, such as the ones involved in sensing chemokines or motogens, could affect cell migration. Alternatively, the electric field could conceivably trigger voltage-gated ion channels or exert forces on adhesion receptors, such as integrins, which result in asymmetrical binding to extracellular matrix (ECM) proteins. Finally, phosphatases, such as Ci-VSP or PTEN, mediate cellular responses to electric fields"

What next ?

When you consider that if what I propose is true the ramifications for neuroscience are there. After all its not just another detail or facet here. Its the entire framework. However its still a backwater project, primarily because firstly that's the only way to develop this. Its not very likely i could develop all these various facets in today's quick results, corporate science career framework.  

So this work itself represents my slow DIY neuroscience education through university part time with the publications being prepped for PhD. This process which demands i produce 3-6 good publications, should get it into the kind of shape to be more usable to apply to neuroscience. 

In the meantime a new area of research has opened up which was predicted by this research. That of EM fields in neurodevelopment. Summary here called

"The bioelectric code: An ancient computational medium for dynamic control of growth and form"

When I get time I will curate the dipole mind project with more skill.  At the moment I am prioritizing on university lab work as well as a commercial mind improvement program.  Some details here