These are alterations of the magnetic fields from sources outside the cranium and outside the myelin sheath which impact the neural processing. Would this not be indicative of quantum influences in neural processing?
They are indicative of EM interference. Quantum interference and effects are something else entirely, and outside of the laboratory, have only been found to occur on very tiny scales associated with atomic structure, as in photosynthesis. Quantum modulation -- that is, a quantum effect that changes the output of a cell based upon evaluation of the inputs -- is possible, but only speculative at this time. Quantum interference, that is, changing the output of a cell based upon external quantum events, is wholly speculative; we're unaware at this time of any such event occurring in nature and I personally, at least, am unaware of a natural means for it to occur.
Given that these effects are sourced outside the cranium, it would seem plausible then that the current generated as a signal propegates[sic] down the axon of neuron A would have an impact on parallel neuron B firing due to the magnetic field generated from A's firing. These generated magnetic fields are strong enough to be detected outside the cranium and are the basis of some FMRI techniques.
FMRI - at least as far as I know - works by intentionally orienting the magnetic impulses of hydrogen atoms, and then uses the newly resulting magnetic field to indicate brain activity by proxy of blood oxygenation. Hemoglobin is diamagnetic when oxygenated but paramagnetic when deoxygenated. This difference in magnetic properties leads to small differences in the MR signal of blood depending on the degree of oxygenation. Since blood oxygenation varies according to the levels of neural activity these differences can be used to detect brain activity. But this is not a magnetic field generated by the brain, it is an externally stimulated one (using extremely powerful external fields) and even so, it does not show any signs of affecting brain function, which in turn argues, again, for the lack of effect of magnetic fields at the level of the neurons.
Here's my thinking on the kind of thing you are talking about, admittedly somewhat off the cuff: Magnetic fields that are generated by current along a conductor are proportional to the inductive impedance of the conductor. Similarly, the amount of current induced in a nearby conductor is proportional to the initial signal size, but reduces by distance (square law) and to the lack of or presence of an impedance match between the two. The poorer the impedance match of the receiving element, the less signal will be impressed upon it by the field, which carries very little power. We must consider that the signals are low and so therefore are the initial field intensities. Because of this, the interior signal condition of the receiving element is extremely likely to drown out -- subsume -- any neighboring interference. And that's actually what FMRI, again the kind I am aware of, indicates. You can boost the magnetic fields within the brain quite a bit and there's no detectable difference in brain function; certainly the brain still works and so I think this tells us pretty clearly that the brain is not very sensitive to this sort of thing. One caveat: magnetic fields generally induce voltages when they are changing, not when they are static, so frequency could easily be an issue here. But now we go back to the experience of high field exposure in the pursuit of various radiative undertakings, again at every frequency from sub-hertz to gigahertz, and this has shown us that the brain continues to operate without any particular notable reaction at all.
Can you give me a pointer to the FMRI techniques you were thinking of?
There are actual articles on inter-neuronal communication via electromagnetic waves: http://www.sciencedaily.com/re... and Neural and Brain Modeling by Rondald MacGregor
These articles talk about electric fields -- not magnetic fields. We do know that the aggregate fields of many neurons firing in relative synchrony can be detected via electrodes, and that creating similar fields internal to the brain affects its operation; but again, we also know from radio and electrical work that the effects of externally applied magnetic fields in inducing electrical impulses are not indicated, probably due to a lack of inductive pickup -- small conductor lengths are only resonant, and therefore capable of efficient power transfer, at extremely high frequencies. In either case -- effect or no -- the much smaller scale (short effective distance) of quantum events doesn't really allow us to apply the information about the one to the other without some guiding evidence that it is actually happening, and that is presently not the case. I don't think it is very likely at all, but as I indicated previously, that's based on the current state of my knowledge about the matter, which I think (hope!) is fairly broad.
Ultimately what this points to is that our mathematical models of neural networks and dynamic bayesian networds[sic] are not exactly what is happening inside the brain. At best its a discrete approximation to a continuous space which exists in a feedback loop with itself. Kinda like a Summation approximation for the Integral of a function.
I am moderately confident that the evidence does not support such a contention at this time. However, let's say that it is so. Then the question becomes, are our current models accurate enough? Given our present ability to reproduce small neural systems and get results that match, it would seem that at small scales -- which is really what we're talking about, I think -- our understanding is sufficient. Certainly from an empirical standpoint, we have been able to do many interesting things with simulation of what I'll call "isolated neural systems" just so we have a handle on the matter, as opposed to an "externally receptive neural systems." Current undertakings show that working to replicate the presently known performance metrics definitely results in usable and very powerful deterministic results. It may well be that even if your suggestion is correct, we won't run into it as a functional limitation although It certainly is interesting to think about even in a speculative manner.
The topological graph structure of the nueron[sic] connections through dendrite and axons is dominant, but it is not dominant enough to eliminate the influence of the fluctuations in the ambient electromagnetic fields. The above articles provide evidence of this. It's not just speculation.
That externally applied electrical fields can create effects is known, and that there are internally generated electrical fields is also known, but the idea that the signal levels of the actions of nearby neurons affect other nearby neurons remains speculation. Even if it is so, and it is significant to how the system processes its signals (see next paragraph), it still falls directly into the electrical, chemical and topological three; none of it signals or suggests the existence of quantum interaction or modulation.
When we create neural networks, we find that they are significantly more useful and effective when the transfer functions are sigmoid or sigmoid-like in that the stable states are high and low, very much binary, whereas the shift between states is not a significant player in stable neural results. I think this is likely also with respect to the neuron (speculating), as neurons that were unable to remain stable would seem to be hard-pressed to represent memory or even consistent thought patterns without disruption. Having said that, however, there are many signals in the brain that operate in repetitive temporal patterns, and perhaps this is a counter indication, though since timing is very much a factor in nervous system communication (see Numenta's work), it may simply be a result of the physical connectivity along the neuron's input channels.
My highest confidence presently rests with the idea that while what we perceive in an aggregate fashion as "mental activity" -- consciousness, memory and so forth -- is both emergent and widely distributed among many neural elements, that distribution is within the network and not any kind of a field effect outside of them, or consequent to significant EM field or quantum interactions between otherwise unconnected neurons. That doesn't mean that applying fields externally won't have an effect -- I'm saying that the brain is apparently stable within its own environment and so modeling, simulation and emulation without non-connected effects is presently what is indicated as the likely equivalent mechanism.