Skip to content

Latest commit

 

History

History
215 lines (136 loc) · 29.8 KB

a0001z.md

File metadata and controls

215 lines (136 loc) · 29.8 KB

a0001z

Self Aware Networks

Years ago online I ran into a group calling themselves Neo Advaita Vedanta, they insisted there was no self. They asked me to look for myself, but to deduce that anything that I found was not actually myself, the goal was to achieve an experiential realization that there was no self, and through that experience someone might experience liberation. I'm not an advocate for Neo Advaita Vedanta but I liked the thought experiment, which was the idea of experientially realizing that there is no real self.

In this thought experiment: try to imagine an organism like a person or an animal from the third person. Imagine how assemblies of cells gain the illusion of having a unified self in the first place.

Imagine how the human brain as a computer system might generate the concept of self, as an internal representation with utility. Imagine how a computer might render or generate it's own sense of self, a character with casual power that represents and is represented by a unified body of causation, defined as having some causal power structure, within some specific volume, ie it's body.

What is the entity that is you? I argue that the "you" is essentially analogous to a computational rendering in a cycling phase oscillation or a branching fractal feedback loop entified via brainwave synchronization of temporal & spatially distributed phase changes across the 3D grid neural-glial network. You are entified via the principles of oscillation, as described in the book Sync by Steven Strogatz. As if each of your neurons is a firefly, or a clock, that is synchronizing with other neurons to act as a single sensor, a single transmitter, a sensor that considers information, via thresholds, and predicts the future, via the principles of memory-prediction (spoken of so well by the folks at Numenta, the book On Intelligence) with reality modeled from synapses to the whole brain with reference frames to help the brain coordinate new incoming sensory inputs (to make sense of the world the brain has a reference frame, a concept I read about in the book A thousand brains) but I am saying the whole brain is doing this at the macro level, not just at the meso level of the cortical columns & the hippocampus, and not just at the level of neurons but also at the level of receptors on basically all cells in the body.

Read the note about Anesthesia decoupling cortical pyramidal neurons in b0232y https://github.com/v5ma/selfawarenetworks/blob/main/b0232y.md

I am a convergence of patterns that exist & oscillate in spacetime. These patterns are rendered by a 3D neural network (the brain) and detected (observed) by the same 3D neural network (the brain) and this process oscillates in spacetime at a rhythm that enables this convergence pattern to happen.

Ripple & Sharp Wave Ripples & Action potential events trigger pond like ripples across the brain.

The exit terminal array of a neuron, when inhibited by the action potential event, represents a magnified pattern to a larger group of cells, a pattern that spreads in an ever larger radius like a stone creating ripples in a pond, because in the later neural array, some other neurons will be exited into phasic bursts as a result of the middle neural array's inhibition, because that third array is recognizing a pattern, and so on.

"Visual evoked feedforward–feedback traveling waves organize neural activity across the cortical hierarchy in mice" "these results provide direct experimental evidence that visual evoked traveling waves percolate through the cerebral cortex and coordinate neuronal activity across broadly distributed networks mediating visual processing" https://www.nature.com/articles/s41467-022-32378-x

"Event-related phase synchronization propagates rapidly across human ventral visual cortex" "Recently, spatial propagation of information has been characterized, in both humans and non-human primates, as travelling waves of cortical activation. These occur at both the macro and micro scale and represent the structured propagation of information across the cortical surface in response to inputs" https://www.sciencedirect.com/science/article/pii/S1053811922003834

"Phase separation of competing memories along the human hippocampal theta rhythm"

"The findings provide evidence that the temporal segregation of memories, orchestrated by slow oscillations, plays a functional role in resolving mnemonic competition by separating and prioritising relevant memories under conditions of high interference" https://www.biorxiv.org/content/10.1101/2022.05.07.490872v1

"Oscillatory dynamics coordinating human frontal networks in support of goal maintenance"

"Responding according to progressively more abstract rules results in increased prefrontal local neuronal population activity (high gamma amplitude, 80-150 Hz) and greater frontal network theta phase encoding (4-8 Hz) which together predict trial-by-trial response times. Theta phase encoding couples with high gamma amplitude during interregional information encoding, suggesting that interregional phase encoding is a mechanism for the dynamic instantiation of complex cognitive functions by frontal cortical subnetworks." "Recent phase/amplitude coupling (PAC) research has found that the phase of low frequency oscillations (e.g., theta; 4-8 Hz) is co-modulated with high gamma activity both at rest and in a behaviorally-relevant manner 18,23-25,27,28,30-33, analogous to evidence that local neuronal spiking activity is biased according to local field potential oscillatory phase (spike/phase coupling)" "These observations suggest a possible mechanism by which spatially segregated neuronal assemblies might coordinate neuronal activity across brain networks29. Specifically, we test a model wherein interregional theta phase encoding coordinates information transfer between frontal subregions during goal-directed behavior38. Such phase encoding would link low frequency phase with high gamma amplitude in a task-dependent manner across phase encoding sites, permitting multiple behavioral goals to be simultaneously maintained"

https://www.researchgate.net/publication/280496530 DOI: 10.1038/nn.4071 

See also note on the Echolocation paper (frontal network oscillations driving vocal cords in bats, an example of top down control with phase wave differentials.) in https://github.com/v5ma/selfawarenetworks/blob/main/a0111z.md

NAPOT 2.0: Individual Receptor Modulation

NAPOT as described in the Whitepaper is correct but the granularity of the transmission between neurons is greater than the Whitepaper draft #3 suggests. This was already reflected in my broader notes but I needed to find more research, and to write something that connected these ideas together more clearly. That's what this note is doing.

The message transmitted between neurons is synapse specific meaning that its not like the neuron is a single led pixel light, which was a thought experiment, but rather it's like the individual synapses on the neuron are the LED pixel lights.

The neurons's exit terminal output therefore is like a mini computer monitor with each synapse being an individual pixel that is modified independently.

So when the neuron fires it transmits a very specific pattern to its exit terminal neighbors.

The precision & granularity of pattern transmission is greater than described with the thought experiment of the exit terminal being like an LED pixel light.

In this expansion of NAPOT every neuron is an even more complex computer, as if the receiving dendrite is the keyboard, and the exit terminal dendrite is the computer monitor with each synapse being a pixel that the next oscillating group of neurons perceives

The idea with sparse distributed representations is that is that you get a 1 when a neuron is active and a 0 when its not

My hypothesis is that the concept might apply to individual synapses increasing the granularity of the rendering of the models within the oscillation activity we call phenomenological consciousness.

I am suggesting that the up regulation & down regulation of individual synapses at the sensory input basel dendrite affects through either the potassium gradient or through cyclic AMP individual calcium channels (up regulation or down regulation of individual calcium channels.) results in a synaptic SDR being output by the exit terminal or the Apical Dendrite.

To connect ideas together, the neuron through the sparse coding of individual synapses, emits a pattern to it's exit terminal array, that is a sparsely coded representation, consisting of synaptic transmissions that are either upregulated or downregulated to trigger excitation or inhibition in the recieving dendrites of the next array. When synaptic backpropagation happens, meaning when the signals are transferred from the receiving dendrite backwards to the exit terminal, what is being adjusted is the synaptic pattern, that represents part of the model of what is being held in the mind, during the oscillating feedback loop of neural activity.

In essence the models of reality, that are oscillating, are alternating between two states, one state is as a patterns held in individual neurons by individual synapses, and the other state is as waves, soliton waves that represent a wave shape with a specific magnitude (amplitude + duration) and frequency pattern (like high frequency bursts or single bursts), and the wave shapes correspond to the synaptic patterns which are like presets that define the wave shapes for when the high phasic action potentials are triggered.

I think there needs to be some clarification about what NAPOT Revision 2 is, and what it is not. NAPOT 2 is not about the individual modification of synapses on the receiving dendrite (or the basal dendrite) and how that relates to potassium, calcium, the Plateau Phase of the action potential duration, and neurotransmitter release. All of that is in NAPOT Revision 1. Instead NAPOT 2 is specifically about the individual synaptic upregulation & downregulation on the exit terminal branches or the Apical Dendrite in the large Pyramidal Cell (referring to Layer 5) STEMMING from receptors & dendritic computations in the Basal Dendrites resulting from sensory inputs to Layer 1.

In short the metaphor I am imagining as a thought experiment is that the Basal Dendrite is a keyboard, whose keys are being pressed at the bottom in layer 1, and typing on that keyboard results in setting up the characters for printing press, putting the letters into place in another section which is layer 5 where the exit terminal or Apical Dendrite is located. When the action potential fires, the message encoded in the printing press of the Apical Dendrite in Layer 5 is printed. The Action Potential is the print function, and the code that is printed is encoded in the EPSP & IPSP of the Exit Terminal (the Apical Dendrite) is what ends up being sent downwards to the Layer 1

Whereas in NAPOT Whitepaper Draft 3 I wasn't sure, or clear on, what was exactly happening to individual synapses in the exit terminal. The conventional idea tossed around is that somehow when the Action Potential fires it transmits an identical signal down all the paths of its exit terminal. Sort of like dividing up the electrical charge of the AP event to each of it's branches. This conceptualization sort of leaves the granularity of computation to either the receiving dendrite with its receptors, or to the neural network at a large scale so that patterns can emergent among many neurons. This is also combined with the conventional idea that the all or none action potential represents a 1 or a 0. NAPOT was addressing that by arguing that because of variation ADP Action Potential Duration there were variation magnitudes of neurotransmitters released with each action potential. NAPOT revision 2 is essentially saying the varying levels of magnitudes of neurotransmitters released happen on a per synapse basis in the exit terminal (or Apical Dendrite) and that means that the pattern a neuron transmits to its exit terminal array (the other neurons) is even more precise. Containing an individual quantity of neurotransmitter release per synapse, and this makes a lot of sense because neurons can release more than one type of neurotransmitter, a distinction that only matters if it is actually used to transmit distinct signals at distinct times.

Essential to the arguement in NAPOT Revision 2 is that the oscillatory & cyclic activity of individual synaptic patterns in the Apical Dendrites with looping activity that involves the Thalamus is allowing the human brain to do Tomography on itself. Consciousness arises in the Feedback loops of our pyramidal cell networks because models of reality are looping in time, rendered in spacetime in the brain, and detected by the brain allowing the brain to do tomography on it's own rendered patterns. The human brain does imaging on itself with oscillatory tomography.

NAPOT Tomography

Someone messaged me saying "Neural array projection and oscillating topography seems to be a merger of neural projection and thalamic rhythms."

The tomography part of NAPOT is how the brain takes patterns that are detected & projected by individual cells and united or renders them into a single pattern with brainwave oscillations.

NAPOT Introduces the concept of computational rendering to neuroscience. NAPOT argues that your brain is doing distributed rendering & distributed observation (detection of rendered patterns) across the whole brain. Allowing our models of reality to oscillate virtually in space & time within the looping grids of brain cells.

Theory of Redness

NAPOT solves the how phenominological consciousness works, how you experience the color red. Red is a phase wave differential pattern, different from the tonic oscillation, so it's a physical contrast to your mind which is also a phase wave pattern. The phase wave of the experience of red as you experience it is rendered by cells in your brain that other brain cells are detecting. This process of pattern rendering & pattern detection is looping for a sustained period of time, and that is the human experience of redness defined.

Oscillating Tomography is like the illusion of the bird in the cage that exists when you have a spinning pattern of a bird on one side of a piece of paper and a cage on the other side of a piece of paper. When the spinning stops the illusion vanishes and you either see a bird or a cage, but the bird in the cage ceases to exist.

Your neurons detect the patterns they are each projecting a small part of because those patterns are tonically oscillating continuously inside the brain like the bird in the cage. Except that the bird in the cage is not a double sided sheet of paper, it's a sparsely coded representation existing across the space of network signals between braincells, with brainwave patterns serving as attractors to unit these patterns into temporal & spatial sequences that are also perceived in a distributed way by the same neural arrays that are rendering (or making) the components of these patterns.

NAPOT Update

A note about NAPOT Revision 2 was added to note a0232z https://github.com/v5ma/selfawarenetworks/blob/main/a0232z.md

"Properties of Sparse Distributed Representations and their Application to Hierarchical Temporal Memory" via Subutai Ahmad and Jeff Hawkins, Numenta

https://arxiv.org/pdf/1503.07469.pdf

This idea is based several things

  1. bidirectional forward & backward synaptic weight changes before and after soma bursts.

a0001z.Milstein (Milstein et al., 2020)

"Bidirectional synaptic plasticity rapidly modifies hippocampal representations"

"behavioral timescale synaptic plasticity (BTSP) can also reshape existing place fields via bidirectional synaptic weight changes that depend on the temporal proximity of plateau potentials to pre-existing place fields. When evoked near an existing place field, plateau potentials induced less synaptic potentiation and more depression, suggesting BTSP might depend inversely on postsynaptic activation. However, manipulations of place cell membrane potential and computational modeling indicated that this anti-correlation actually results from a dependence on current synaptic weight such that weak inputs potentiate and strong inputs depress. " "The core feature of such plasticity mechanisms is that they are autonomously driven by repeated synchronous activity between synaptically connected neurons, which results in either increases or decreases in synaptic strength depending on the exact temporal coincidence" "plateau potentials acting as the delayed factor that converts synaptic ETs into changes in synaptic strength." "However, BTSP was shown to strengthen many synaptic inputs whose activation did not 60 coincide with any postsynaptic spiking or even subthreshold depolarization detected at the soma (13), suggesting that changes in synaptic weight might be independent of correlated pre- and postsynaptic activity, and that BTSP may be fundamentally different than all variants of Hebbian synaptic plasticity" "We found that dendritic plateau potentials rapidly translocate the place field position of hippocampal place cells, both by 70 strengthening inputs [synapses] active near the plateau position, and weakening inputs [synapses] active within the original place field." "In order to determine if the increased post-synaptic activity in place cells is causally related to the synaptic depression observed within the initial place field, we performed a series of voltage perturbation experiments, which indicated that the direction of plasticity induced by plateau potentials is independent of post-synaptic depolarization and spiking." "Next, we inferred from the data a computational model of the synaptic learning rule underlying this bidirectional form of plasticity, which suggested that it is instead the current weight of each synaptic input that controls the direction of plasticity such that weak inputs potentiate and strong inputs depress."

"In most cases the evoked dendritic plateaus shifted the location of the neuron’s pre-existing place field towards the position of the second induction site"

"The world is full of obvious things which nobody by any chance ever observes." Sherlock Holmes.

I imagine that the induction site of the initial place field is like marking on a rotating cyclinder, or gear, in a clock mechanism, and the second induction site, is like shifting the marking.

This validates the NAPOT Revision 2 concept I have that when burst firing happens, the action potential is like the print function, and the synaptic output to the next neural array is the printed message. The message consists of up or down regulated synapses that were set prior to the burst firing event. This rendering is the projection in NAPOT theory: Neural Array (Projection is the Rendering) Oscillation Tomography. Or Neural Array Synaptic Projection Rendering Activated by Soma bursting and looped in cortical cortical array feedback loops and cortical thalamic array feedback loops.

https://sciety.org/articles/activity/10.1101/2020.02.04.934182 full article https://www.biorxiv.org/content/10.1101/2020.02.04.934182v2

  1. The idea that metabotropic receptors on the receiving dendrite might trigger cAMP messages to change individual calcium channels in the exit terminal. The fact that synapses are individually inhibited or excited points to this possibility that the neurons output to its network is tailored synapse by synapse to represent memories with greater resolution & definition. See note a0272z on cyclic Amp modulation of specific ion channels https://github.com/v5ma/selfawarenetworks/blob/main/a0272z.md

  2. Sparse Distributed Memory by Pentti Kanerva: The idea of individual synaptic modification is consistent with the concept of Sparse Distributed Representation (a major topic at Numenta), meaning that only a 2 or 3 of the exit terminal synapses need to be excited, and the rest inhibited, for one neuron to pass on a very specific pattern to its exit terminal array.

The concept Reference Frames may apply

The idea of "References Frames" in our Cortical Columns from the book A Thousand Brains may apply to the slow tonic oscillations that are brain wide, in relationship to the incoming high phasic signals, or soma burst action potentials.

The higher layers of the brain, the cortex, and the thalami, tend to have slower delta & theta frequencies with a higher magnitude lower frequency. While the incoming sensory inputs to the lower layers of the brain, the cortex, and the thalami seem to trend with higher frequency brainwaves: alpha, beta, and gamma frequences. Although I have read some papers that seem to contradict this hypothesis at times, more research is needed.

In general I think there is evidence that the Hippocampus is essentially a special cortical column, like Cortical Column #1, after reading Buszaki 2006 I think that should be commonly accepted.

Self regulating receptor channel? say what?

"Researchers identify a new mechanism responsible for controlling auditory sensitivity"

"a newly identified mechanism of how auditory sensitivity is regulated that could temporarily reduce sensitivity of the auditory system to protect itself from loud sounds that can cause irreversible damage. The study, led by CU Anschutz researchers Andrew Mecca and Giusy Caprara in the laboratory of Anthony Peng, tested a decades-old hypothesis which proposed that the gating spring, a tiny, nanometer-scale protein structure which mechanically opens and closes an ion channel in sensory hair cell cells in response to sound vibrations, can act directly as a controller of the channel's activity. Previous work in the auditory field has focused mostly on understanding mechanisms which target the ion channel. This study provides the first evidence that the gating spring itself has the capacity to modulate the sensitivity of the channel." I didn't hear you. https://medicalxpress.com/news/2022-07-mechanism-responsible-auditory-sensitivity.html?fbclid=IwAR1FiPuw4bMNjPhKuEIp_CyAWWLF2cUdy0pvlgA4rCinmoBPdZUdxDsqmyg

my hunch is that this paper (below on Auditory...) is chasing the wrong hypothesis because the major effect of calcium spikes is not going be on the dendritic side but rather on the exit terminal side of the neuron. I want to look up the sources of their hypothesis that they tested to understand if they interpreted the research incorrectly

"Auditory corticofugal neurons transmit non-auditory signals to support discriminative learning"

https://www.biorxiv.org/content/10.1101/2022.08.08.503214v1

"Complexity in Searching for the Neural Code"

https://jonlieffmd.com/blog/complexity-in-searching-for-the-neural-code

(Next one below related to Alpha Oscillations being relevant for rendering the mind, Neural Oscillatory Tomography)

"Alpha oscillations shape sensory representation and perceptual sensitivity"

"Here, we show that ongoing alpha-band activity in occipital-parietal regions predicts the quality of visual information decodable in neural activity patterns, and subsequently human observer’s sensitivity in a visual detection task. Our results provide comprehensive evidence that visual representation is modulated by ongoing alpha-band activity" https://www.biorxiv.org/content/10.1101/2021.02.02.429418v2#review

(Saving this next article below for an attention schema (Gazzaniga) map)

"Awareness-dependent normalization framework of visual bottom-up attention"

"Our findings indicate an awareness-dependent normalization framework of visual bottom-up attention, placing a necessary constraint, namely, awareness, on our understanding of the neural computations underlying visual attention." https://www.biorxiv.org/content/10.1101/2021.04.18.440351v1#review

(oscillat, render, field, graph, causation, emotion, tomography) Not Holography, Mindography, or the rendering of consciousness moment by moment, as a reaction to the universe, learned by cells, the process of natural selection

3D Neural Networks (Point Cloud Semantic Segmentation, Denoising, Real Time Raytracing, and Self-Aware Rendering) you are the rendering that your brain is rendering (oscillat, synap) map: synapse configuration, SDR, Oscillator, internal recognition

The idea that a neuron's synaptic configuration represents a lot of different SDR patterns, that emerge from combinations of neural firing into distinct internal renderings that an oscillator can recognize

Past examples of people using EEG and or MRI to translate images from brain activity involve sorting large scale noisy signal data into neural networks to predict or guess what picture the brain could be seeing based on an existing set of images that the researchers pick, however that its not the same level of brain machine interface guess work as generating images movies and narratives from neural activity alone without researchers supplying categories of images themselves.

Because oscillations transmit everywhere in the brain any part of the brain can serve as a reader for saving all of the brain's oscillatory patterns.

A new method for how to download the entire contents of the human brain into a computer.

Essentially the computer can do neural imaging tomography to create a 3D map of the oscillating structure of the human mind, predicting the entire oscillating structure with neural oscillation tomography.

It's like recording a 3D movie of all your brain activity, from which the computer can infer the changes in all your neural connections,

I think people think that the oscillations are just about timing your responses, or the temporal cadence of your thoughts, but I'm suggesting that the timing is shaping the internal representation, so that the map of reality is rendered just in time with muscle coordination speed.

The real speed of cognition is about the number of simultaneous operations I think, not about the temporal cadence of the sensory field, and the representation of the main body of causation (the human muscles)

The brain introduces its own timing delays to coordinate its predictions to align reality between what you see & hear. To create a perception of reality that seems whole from lots of little sensors that make up your eyes, ears, nose, tongue, skin, and body.

The idea is that you are the rendered multi-sensory perspective and the model of your body of muscular causation, and this includes the thought stream, the sequences of patterns, emotions, feelings, these are all part of the sensor-motor model that the whole brain is making

You, the rendering, and it, the chemical structure oscillating with the phase field of space are sharing each interval of the moment that is being rendered as the organism's map of reality. You, the virtual entity of information existing as phases (frequencies) are sharing this moment with your saltwater, protein, chemical body, like two halves of one oscillating nexus.

The mind itself is like a rendering, even the non-visual parts, in that a rendering is a frame in a movie, or a moment in a sound track, that is made up of a lot of different parts brought together to make that one frame. In each interval of time your brain state, and your brain waves representing one facet of your brain state are changing to represent that next moment in time as the next rendered frame. At the micro level that consists of receptors that fired, neurons that fired, neurons that were inhibited, and other states that cells, and parts of cells can be in. At the meso level we are exploring changes to neural circuits, and cortical columns, and at the macro level it is about studying connections between major regions of the brain, the thalamus, the hippocampus, the entorhinal cortex, the tempo-parietal junction, the rich clubs, the default mode network, the connections between pyramidal cells and the thalamus and so on.

Read "Closing the mechanistic gap: the value of microarchitecture in understanding cognitive networks"

"default mode network (DMN) and multiple demand network (MDN)] underpins their broad involvements in cognition." "Recent work has established that the activity of brain areas often varies in concert, giving rise to the notion of functional brain networks [2.], as well as to the associated hypothesis that the relationship between brain activity and complex behaviours is best operationalised by the interaction of distributed nodes that constitute these networks [3.,4.]. " "We suggest that a brain-based interpretation of human cognition can be enriched and guided by studying the microarchitecture of large-scale functional networks, notably the DMN and MDN." "In particular, our review highlights how microarchitectural heterogeneity impacts on brain function across scales: (i) local cytoarchitectural variations within subregions echo motifs of functional organisation across the brain [54.], (ii) microarchitecture provides complementary information to structural connectivity in predicting functional connectivity [53.,55.,83.], and (iii) the DMN and MDN are variably spread across large-scale axes of microarchitecture and function" https://www.cell.com/trends/cognitive-sciences/fulltext/S1364-6613(22)00158-9?dgcid=raven_jbs_aip_email#.YuTn1ADdYqY.twitter

Read more about memories scaling up in note a0126z

This article below has conclusions that are very similarities to some high level conclusions in my work, but I have published my work on github & social media. I don't have to worry. Their argument also has enough differences to be a citation in my work, that seconds some high level conclusions in my work, my work that predates their paper.

"A Relativistic Theory of Consciousness"

https://www.frontiersin.org/articles/10.3389/fpsyg.2021.704270/full

After reviewing "A Relativistic Theory of Consciousness" a second time, it turns out the similarities to my work are much more superficial or surface level than I realized this morning, it shares a few high level conclusions. It's essentially a philosophy paper with none of the underlying neuroscience. No worries. I did like reading their paper. I respect it for what it is. I recommend reading it.