Context effects produced by question orders reveal,quantum nature of human judgments

Hi,

DNA and other systems in the cell are used for learning and memory, as in single celled creatures, since they don't have a neural network brain, they have to use other structures in the cell for learning and memory. This includes the membrane proteins, and basically the whole cell when you really look at it, is learning/reacting all the time to make the smartest choice. For example the choice to turn into a spore is extremely complex, as ideally the cell will not want to do this ever as it means a stop to its reproduction and an evolutionary disadvantage when it wakes up. So even in difficult to survive situations the cell will try to keep from going into spore form but if it doesn't turn into a spore at some point conditions can become bad enough that it is too late to turn into a spore. The cell figures out when to turn into a spore better than a human thinking about it could still probably!

Single celled organisms can even have their own immune systems:

If you assume these similar cellular abilities are active in more complex organisms like humans, then it is far more complicated than a simulated neuron. Also the idea that real neurons are actually cells with no abstraction as simulated neurons have, this is an important difference. People who think they can recreate real things with an abstraction model have a lot to prove since it is a big assumption.

This seems to be a good place where self-awareness could reside as the lack of any abstraction layer in a mind bogglingly complex system.

cheers, Jamie

Hi,

Artificial neural networks that dream? Do you have a link? I googled but didn't see anything, maybe I was asleep.

cheers, Jamie

yep - neural nets are a handy software tool with only a vague resemblance to an actual neuron.

on the QM topic: Where the Quantum mechanics comes in, in the synapse, is possibly the ion channel for the neurotransmitter. (serotonin?) Evidently the channel is about

But to my knowledge there is no theory as to how a localized quantum effect in a synapse could maintain coherence in a macroscopic scale. Perhaps non-quantum axons and quantum at the junctions? Quantum effects in the axon? (axon = transmission wire between neuron bodies.)

WOW.

Hi,

Maybe the macroscopic coherence that seems impossible to extend beyond the small scale synapse is actually between the brain (on all scales) and the Schrodinger equation of the universe (that is a coherent quantum system) :D

cheers, Jamie

DNA and mitochondria within a cell are probably non-trivial quantum computers. Nature seems to optimize the use of anything that's not physically forbidden.

Photosynthesis uses explicit quantum effects, so why not neurons?

Thanks. I must admit I do not grok QM enough to have insight about DNA being a quantum computer. Nor Jamie's interesting post about interaction with the universe.

That English physicist may have an overall theory of it - the quantum computing exponent. (name on tip of tongue.)

Hi,

The idea of "computing" per say usually implies a form of abstraction or symbols, it used to be a bit popular to see biological processes as forms of computation, but I think the computer science people shot it down. Every process in the organism has been created to maximize survivability. The way the brain operates is arguable computing though since it is a layer of abstraction and symbols on top of biology, with variable goals. Low level biological systems can be used to create computing though, ie DNA computing:

cheers, Jamie

Hi,

A neat thing about the quantum coherence effect in photosynthesis is that in low light conditions where very little energy is obtained from the light it is apparently more energy efficient to let the quantum coherence collapse. It must take energy to maintain the quantum coherence in the chloroplasts so if the light intensity is low enough then it turns off even though the light to chemical energy pathway is less efficient with the quantum coherence turned off (more electron resistance or something through the chloroplast).

This also is nothing to do about computation as far as I know, it is just like the difference between a copper wire and a superconductor (same idea).

cheers, Jamie

no faith in the free market?

tomorrow it would, see argument 1.

That is a fair point at the ultimately smallest atomic and molecular scale but it is far from clear that it affects macroscopically sized objects like neurons and synapses. They are big enough to behave like idealised classical components to a very good approximation.

I can believe that quantum mechanics is what makes DNA a four base quantum binary code since in QM a single compare matches one of four.

And even the slightly more controversial conjecture that the DNA codons are 3 bases long for a library of ~20 amino acids because that is about the limit for a QM three step comparison (ISTR 23) and the two base limit (ISTR 12?) isn't quite enough to do anything interesting.

Even our classical silicon computers are not quite deterministic if a suitably energetic cosmic ray flips a bit in one of their memory locations although they tend not to recover so gracefully as a brain. More of a problem in space flight than for terrestrial computers.

Only a major factor on the very smallest scales. The largest thing that has been quantum mechanically diffracted is a bucky ball. Neurons are very much larger than that. Their internal chemistry is quantum mechanical in nature but their external properties are to all intents an purpose classical. The problems are in instrumenting them.

Horseshoe crabs were popular for this research as their primitive eyes rods, cones and neurons are exceptionally large. Trilobite on steroids.

Here is a paper. Not the one I had in mind which I think was from IBM at about the same time as the cat brain simulation announcement/claim.

The networks self organise in much the same way as as a CA does.

Stephen Wolfram was at one time trying to promote the idea that the entire universe is one big CA (and conversely that you can simulate anything using a suitably chosen CA). Turing machine being one such.

That depends on the neural networks. The toy ones sold to be trendy solutions to tricky problems I would have to agree with you, but there are pretty much complete simulations of neurons that reproduce the observed potential waveforms at synapses too. I doubt veyr much that it is necessary to simulate a neuron in infinite detail to make a working computation model of a human brain.

Meaningless crank word salad. It would make just as much sense after Shannonising.

The four base DNA/RNA code is the quantum mechanical equivalent of binary logic. You get a four way matching in a single comparison.

I have the details somewhere for nested quantum compares and ISTR the limits are something like 12 and 23 respectively (I could look it up). It is a conjecture that the 3 base codons are because QM makes that the most efficient way to encode enough amino acids to make proteins for a self organising system to arise mediated by messenger RNA.

The mitochondria are the energy organelles and as such burn fuel to power everything else. Energy transport in cells is largely chemical in nature although the efficiency is improved by quantum coherence.

All chemistry at a molecular scale uses explicit quantum effects. Photosynthesis utilises coherence and resonance to make the energy transfer more efficient but that is on a molecular scale.

Fire uses explicit quantum effects too. Do you think it is self aware?

Neurons are too large to be experiencing strong quantum effects.

Hi,

Here's a video on what is actually going on inside a real synapse, it is extremely complex, nothing close can be made by humans for a long time.

cheers, Jamie

Hi,

Here's a video on what is actually going on inside a real synapse, it is extremely complex, nothing close can be made by humans for a long time.

oops forgot the link:

cheers, Jamie

Oh goody! Random word salad that is invariant under the Shannoniser.

If that is a proof then I am Alexander The Great!

Not everybody feels pain. Although the lack of pain receptors is a very dangerous sensory defect it does occur and ditto for each of the other senses. Losing sense of taste or smell are the least damaging in the modern era although in past times they were essential for survival.

The Godel incompleteness theorem isn't relevant to qualia.

Giving something a name does not make it real or useful unless you can use the new construct to provide a greater insight into the subject.

That is defeatist in the extreme. I am fairly sure that a sufficiently large neural network with appropriate inputs will at a certain scale and beyond become self aware just like we are. That will pose serious ethical questions when it happens not all that long into the future.

Dennett thinks clearly - you do not.

Hi,

Just because a system of particles can't be diffracted doesn't mean that there isn't quantum uncertainty in them individually. All particles have uncertainty as a fundamental part of what they are.

Ya the internal part of the neuron is more complex than the external properties, but I think it is a volume to surface area ratio of complexity, as the external properties all pass through a 2D membrane, but the internal chemistry is 3D and has more volume. It is all still filled with quantum uncertainty though. The actual synapse firing is even more complex than most people say, ie: "noise miniature neurotransmission" is actually a part of the brain firing that is usually not considered:

cheers, Jamie

Hi,

Thanks, it has no mention about dreaming, but talks about "Hierarchical Modular Networks" HMN's, and their "self organized criticality"

from the paper: "We found that a hierarchical modular architecture can generate sustained activity better than random networks"

A random network is like an un-programmed FPGA, the modules are what gives function to the network. The self organized criticality (SOC) is basically the modules interacting with each other, and apparently is "a widespread property of complex systems, such as piling of granular media, earthquakes, and forest fires".

Sure the universe is one big cellular automaton (CA) described by a Schrodinger wave equation maybe.

"In lattice QCD, fields representing quarks are defined at lattice sites (which leads to fermion doubling), while the gluon fields are defined on the links connecting neighboring sites. This approximation approaches continuum QCD as the spacing between lattice sites is reduced to zero. Because the computational cost of numerical simulations can increase dramatically as the lattice spacing decreases, results are often extrapolated to a = 0 by repeated calculations at different lattice spacings a that are large enough to be tractable."

Good luck ever getting a discrete computer that can simulate with a lattice spacing of zero which would be required unless space is quantized, pretty sure it is not, if you assume it is then paradoxes are everywhere.

cheers, Jamie

We don't have to compute every detail of the internal workings of a synapse any more than track every molecule of air to compute the windspeed. It is sufficient to model the potentials on the membranes and some pretty good PDEs already exist for doing this.

This isn't a bad introduction suitable for electronics engineers on single neuronal modelling :