a0007z
Dendritic morphologies & synaptic connections represent learned memories. Memories activate when preferred patterns (the memory) are detected, this causes cells to create a selective transmission or a projection to other cells (neural arrays). In this note I am looking at Layer 6 & Layer 5 Pyramidal cells, to learn how the Apical Dendrite, or the Exit Terminal, is contributing to the rendered projection of that learned memory to the rest of its network. The main idea that I am exploring is that the signal transmitted from the exit terminal is not uniform along all branches, but instead is modified synapses by synapse, and that the memories in the exit terminal also transmit backwards to play a role in multiple seconds long coincidence detections involving back propagating action potentials from the exit terminal or Apical Dendrite to the Soma, that when combined with the Basal or sensory input dendrite can cause a coincidence detection that creates a special type of action potential event called burst firing.
"All pyramidal neurons so far investigated have displayed active propagation of action potentials (APs), from the soma along the apical dendrite supported by voltage-gated dendritic Na+ channels (Spruston et al., 1995; Stuart et al., 1997; Waters et al., 2003), modulated by dendritic K+ channels (Bekkers, 2000; Johnston et al., 2000; Korngreen and Sakmann, 2000; Schaefer et al., 2007), and accompanied by influx of Ca2+ ions (Markram et al., 1995; Larkum et al., 1999a; Barth et al., 2008). Another prominent feature of pyramidal neurons is the ability of the apical dendrite to generate local spikes with voltage-gated Na+ and Ca2+ channels (Kim and Connors, 1993; Schiller et al., 1997; Golding et al., 2002; Gasparini et al., 2004) as well as NMDA receptor channels (Schiller et al., 2000; Larkum et al., 2009)." https://www.jneurosci.org/content/30/39/13031
"Layer 6 (L6) pyramidal neurons are the only neocortical pyramidal cell type whose apical dendrite terminates in layer 4 rather than layer
- Like layer 5 pyramidal neurons, they participate in a feedback loop with the thalamus and project to other cortical areas."
"We found that L6 pyramidal neurons share many fundamental dendritic properties with other neocortical pyramidal neurons, including
-
generation of local dendritic spikes under the control of dendritic inhibition "Inhibitory control of calcium electrogenesis. Dendritic inhibition has been shown to powerfully block dendritic Ca2+ electrogenesis in neocortical and hippocampal pyramidal neurons (Buzsa´ki et al., 1996; Miles et al., 1996; Pe´rez-Garci et al., 2006; Larkum et al., 2007; Murayama et al., 2009)."
-
voltage-dependent support of backpropagating action potentials
-
timing-dependent dendritic integration
-
distally located Ih channels
-
frequency-dependent Ca2+ spike activation
-
NMDA spike electrogenesis in the distal apical dendrite.
"Corticocortical versus corticothalamic neurons. L6 pyramidal neurons can be divided into two categories based on the projection of their axonal arborizations: corticothalamic (CT) and corticocortical (CC) projecting neurons (Zhang and Desche`nes, 1997; Kumar and"
https://doi.org/10.1523/JNEUROSCI.2254-10.2010
"Physiology of Layer 5 Pyramidal Neurons in Mouse Primary Visual Cortex: Coincidence Detection through Bursting" 2015
"L5 pyramidal neurons are the only neocortical cell type with dendrites reaching all six layers of cortex, casting them as one of the main integrators in the cortical column."
"100 tuft and 175 basal NMDA/AMPA synapses are distributed randomly across the apical tuft and basal dendrites of a multi-compartmental L5 pyramidal neuron model." "(b) Simultaneous tuft and basal inputs triggers a burst of somatic APs and a dendritic Ca2+ spike "(c) basal inputs alone evoke only a single" somatic spike." "(d) Apical tuft inputs alone do not evoke somatic spiking"
An interesting thought about the cyclic nature of Layer 5 pyramidal cells, is the idea that the process of the dendrite being a sensor, like an eyeball or an ear appears to apply to both sides of the neuron, on the way in, and on the way out. So the Apical dendrite, as a neural network, is learning from incoming sensory stimulus, but the Soma burst is one of the incoming sensory stimuli, with models of soma burst waveforms becoming part of what the Apical Dendrite has to contend with.
I think of it as like the basal dendrite is learning to respond selectively to certain patterns, the apical is learning what patterns it responds to.
My intuition is that the apical is learning its own cell's response, and through feedback loops, with itself and with other pyramidal neurons the cell is going to make predictions about its own patterns of causation, what patterns it responds & how it reacts or responds. With the Apical Dendrite perhaps symbolizing a more complex response that is connected to more complex learning. (compared to other types of neurons)
The fact that bursting (bursts of somatic AP) results from the coincident activity of basal + apical, but not from either one separately is a different coincidence detection mechanism from the leaky integrate-and fire model of a neuron which explores coincident synaptic inputs, leading to coincident dendritic activity, which leads to the phasic action potential from just the south side of the Soma.
In the leaky integrate & fire neurons, and the Hodgkin & Huxley Model there is very little thought to the concept of the Soma burst spiking being a coincidence detection from both sides of the neuron. But if it's true for Layer 5 pyramidal cells, then we ought to know if there is a greater role for the exit terminal in all other types of neurons.
So perhaps we can think of the axon, with it's different types of action potentials, different durations (wave shapes, magnitudes), and different burst rates, as a analogous to a lens, a lense that is creating a summation signal that the exit terminal is learning as it's learned response.
The idea is that our sensory inputs are sort of encoded in the input dendrite (or the basal dendrite) and our motor outputs are sort of encoded in the exit terminal (the Apical dendrite), and through cyclic looping activity our input learning (sensory in the bottom dendrite) is being mapped to our output learning (motor in the output exit terminal, or the Apical Dendrite)
So every neuron represents an actor, one that learns to respond selectively to certain patterns, and encodes in its exit a response action.
The meaning of the main somatic signal is like the summary pattern of what the dendrite (basal) has detected, a magnitude + a frequency rate, and the exit or Apical is more about having a highly coordinated response.
https://github.com/v5ma/selfawarenetworks/blob/main/a0232z.md
It's interesting that a place field, in place cells, can begin with a somatic spike. (see note b0232y on behavioral time synaptic plasticity) but it results in burst firing the next time the organism encounters the same place. Perhaps that first spike comes from the Dendrite, and then the Exit Terminal learns to predict it, based on the wave shape (duration & frequency), so that the exit terminal is sending it's prediction backwards, a back propagation spike, at the same time that the mouse encounters that same place again. That would make both sides of the neuron like timers, timers that learn when they are going to fire, and the dendrite is learning a pattern, encoding it in a timer signal at the axon, with the exit terminal learning the different timing signals, and responding with its own backpropagating signals, to create burst spiking that leads to an organism taking actions based on it's memories.
"This line of research has shown that pyramidal neurons in different cortical regions contain voltage-gated Na+ channels along the dendritic trunk which support the backpropagation of action potentials (APs) from the soma into dendrites [6], as well as voltage-gated Ca2+ channels that support spiking in the apical dendrite"
"Using a detailed multi-compartmental model, we show this physiological setup to be well suited for coincidence detection between basal and apical tuft inputs by controlling the frequency of spike output. We further show how direct inhibition of calcium channels in the dendrites modulates such coincidence detection."
"Because past studies have found that dendritic electrogenesis depends on calcium channels in the apical dendrites, and that sag and dendritic resting potential and input resistance is dependent on Ih conductance, we were able to manipulate only dendritic calcium and Ih conductance parameters, by hand, to fit to our experimental results."
- Coincidence detection between basal & apical dendrites via the frequency of the spike output.
- Synaptic Efficacy: Inhibiting calcium channels modulates the coincidence detection between the basal & apical dendrites.
- This is going to allow for multi-second pattern learning. Or Behavioral Timescale Synaptic Plasticity (BTSP) in the cortical columns of the neocortex. I share notes about the documentation of BTSP in the hippocampus in note a0232z https://github.com/v5ma/selfawarenetworks/blob/main/a0232z.md
I've read about back propagating action potentials from the postsynaptic membrane (traveling backwards) to the presynaptic membrane. That's where the idea of individual synapses being up regulated or down regulated might be like setting up the individual letters of a printing press (the old fashioned kind that prints newspapers) with the action potential event printing that pattern to its exit terminal network, or the receiving neural array.
"We find that mouse L5 pyramidal neurons in V1 support backpropagating action potentials (bAPs), and dendritic Ca2+ spiking."
However the combination of the research on BTSP in the Hippocampus, with back propagating action potentials from the Apical dendrite towards the soma in Cortical Columns is for me a new mechanism of coincidence detection across the whole brain, and it ought to lead to studies that explore whether much smaller neurons also have burst spiking that is triggered by the combination of back propagating signals from the exit terminal direction + forward propagating signals from the dendrite that represents sensory input. This is not the Hodgkin & Huxley model of neuron. It's not the leaky integrate & fire neuron, or perhaps I ought to say it is more than the classic model of neural firing.
"We show how concurrent input into the perisomatic and electrically remote apical tuft regions switches somatic action potential output from low to high frequency. Thus, high frequency bursting indicates coincident input into different parts of the dendritic structure."
"We further explore regulation of this coincidence detection scheme by blockage of the voltage-gated calcium channels on which this computational principle depends. We ultimately show that such single-cell computation can be conceptualized as a composite of sigmoid functions."
"A phenomenological model" "To conceptually address the single-neuron computation that this biophysical setup performs, we establish a phenomenological model. We compare three models: a composite, multiplicative, and an additive model" "Each of these models uses two sigmoidal functions to perform intermediate computations, and is justified by the existence of the two separate (one dendritic and one somatic) spiking zones in the neuron." "The interaction of the sigmoids in the composite model is justified by the experimental results suggesting that the result of dendritic electrogenesis is to lower the threshold for a high-frequency burst at the soma." "The multiplicative and additive models have two independent sigmoids, one of which takes tuft (apical) synapses as input, and the other which takes basal synapses as input. " "The composite model outperforms both the multiplicative and additive models, though less so when Ca2+ conductance is decreased by 75%, suggesting that the inability of the multiplicative and additive models to represent the input-output relationship depends on dendritic electrogenesis" "The effect of tuft input (The Apical Dendrite or Exit Terminal) is to increase the maximum possible output frequency (Fig. 5D left) and decrease the threshold of basal input needed to elicit high frequency firing (Fig. 5D right)." "Thus, the sigmoid that relates tuft input to burst-firing threshold is decreasing (since more tuft input decreases that threshold, Fig. 5A red sigmoid), while the sigmoid that relates tuft input to somatic output frequency is increasing (since more tuft input increases the output frequency, Fig. 5A blue sigmoid)." "The composite model describes a coincidence detector between basal and tuft input, since only when both input streams are active in sufficient amounts is the resultant output high frequency."
Should we multiply two sigmoids (from both sides of the neuron) together, should we add two sigmoids each representing (inputs from each of two sides of the neuron together, or should we have a single sigmoid that is modified on each end?
"The composite model describes a coincidence detector between basal and tuft input, since only when both input streams are active in sufficient amounts is the resultant output high frequency."
"How might such single cell computation be involved in visual processing? To explore tuning properties of cells employing a variety of mechanisms, we used circular distributions (von Mises distributions, see Methods and Fig. 6 and S5) to model inputs as a function of stimulus orientation. We compared four different mechanisms (Fig. 6A). A composite sigmoid as described previously, a purely multiplicative where the number of tuft and basal inputs are simply multiplied to arrive at output, a purely additive, where the number of tuft and basal inputs are simply added to arrive at output, and a single sigmoid mechanism, where either the tuft or the basal input is put through a sigmoid function to arrive at the output."
See Figure 5 (save to pictures folder)
"Coincidence Detection in Pyramidal Neurons Is Tuned by Their Dendritic Branching Pattern" https://doi.org/10.1152/jn.00046.2003.
This study (link & name above) backs my assertion that a memory is both the synaptic connections, and the dendritic morphology (an assertion found in the NAPOT Whitepaper as well as other places.)
(from the above study) "Is morphological variation a bug that arises from intrinsic biological variability or a feature that extends the computational capabilities of a neuron?"
(from the above study) "We conclude that variation in dendritic arborization may be a key determinant of variability in coupling (49 17%; range 19–83%; n 37) and is likely to outweigh the contribution made by variations in active membrane properties. Thus coincidence detection of inputs arriving from different cortical layers is strongly regulated by differences in dendritic arborization."
"Here we focused on variations of the dendritic branching pattern within one class of cells, thick tufted neocortical layer 5 (L5) pyramidal neurons. This cell type is one of the largest found in the mammalian cortex. Its prominent apical dendrite extends vertically 1.3 mm from the soma and gives rise to a tuft of dendritic branches in layer 1 and oblique dendrites spreading horizontally in layers 2/3 and"
"4. The detailed branching pattern varies considerably between individual cells. L5 pyramidal cells also display 2 major types of regenerative events: First, sodium based APs (Na+-APs) are initiated close to the soma and propagate back into the dendritic arbor [backpropagating AP (bAP); Stuart and Sakmann 1994];"
"second, long-lasting, mainly Ca2+ mediated depolarizations (Ca2+-APs) are initiated predominantly in the distal regions of the apical dendrite (Helmchen et al. 1999; Schiller et al. 1995, 1997) and can be evoked by strong distal synaptic input. It was recently shown that a bAP can lower the threshold for initiation of a Ca2+-AP, thus enabling L5 pyramidal neurons to couple synaptic inputs from different cortical layers if they coincide within a short time window [Fig. 1; backpropagation activated Ca2+ spike firing (BAC firing); Larkum et al. 1999a,b, 2001]."
Dendritic arborization & branching (or dendritic morphology, including between Basal & Apical Dendrites) is one of the ways neurons selectively tune into certain criteria for memory-prediction while ignoring other criteria (the components of patterns). Individual parts of the dendrites & individual synapses are able to excite (up regulate) or inhibit (down regulate) based on coincident detections across the cell membrane, thus allowing the brain to carefully define its own internal mental models of reality, and to actually fine tune those mental models of reality.
The article documents an effort to model or simulate dendritic morphology.
"Dendritic fingerprint captures variability of simulated coupling. The dependence on dendritic arborization in simulations suggests that it should be possible to extract geometric parameters that contribute to this variability."
"a more detailed dendritic fingerprint (Fig. 4) of L5 pyramidal neurons: it consisted of 1) one- and 2) two-dimensional branching densities, 3) the width of the dendritic tree, and 4) diameter distributions for the apical and 5) basal dendrites, as well as 6) the number of basal dendrites, 7) somatic diameter and length, and 8) the maximal length of the non normalized apical dendrite."
I'm surprised there was not more emphasis on the lengths of each dendritic branch, because one of the major differences between Mice & Human brains is the length of the dendrite. (search this note for lengthreference1)
"The most profound difference in the branching parameters of the strongly and weakly coupling cells was in the branching densities in the tuft"
This goes back to the question of how a ganglion neuron in the eyeball is able to selectively respond to only patterns that move from left to right, and not patterns that move from right to left or from top to bottom or bottom to top for example. The dendritic morphology means that patterns triggered from the left to the right arrive together in time to trigger that neuron to fire. That is what led me to the concept of HOW the dendritic morphology, along with synaptic connections, both their location & quantity define LTP in the physical morphology of the cell. This concept contributed to my "Neurons are transmitting their shapes'' theory in 2014 that I shared on social media via an alias, but you can find in not a0008z https://github.com/v5ma/selfawarenetworks/blob/main/a0008z.md
"Visual Responses of Ganglion Cells" https://www.ncbi.nlm.nih.gov/books/NBK11550/
"Different cells become selectively tuned to detect surprisingly subtle "features" of the visual scene, including color, size, and direction and speed of motion. These are called "trigger features"."
On "Contrast Sensitivity Functions and the Difference of Gaussians Receptive Field Model"
""Contrast" is the difference in brightness between the light and dark phases of the pattern. In this receptive field characterization, contrast is reduced until the ganglion cell just barely responds to the introduction of such a stimulus into a featureless field. This is the threshold. The process is then repeated for patterns with different bar widths, or "spatial frequencies" (the reciprocal of the number of bright- and dark bar-pairs per unit distance). The sensitivities (the reciprocals of contrast thresholds) are plotted as a function of spatial frequency. The net result is a curve (Fig. 8), which demonstrates that in ganglion cells there is an optimal spatial frequency of stimulation."
On "Directionally Selective Ganglion Cells" "Directionally selective retinal ganglion cells respond to stimuli moving in a preferred direction and are inhibited by stimuli moving in the opposite or null direction." "In addition to a preferred direction of movement these cells have preferred speeds of movement, some prefer slow movements, whereas others prefer rapid movements." "Directionally selective ganglion cells, stained after microelectrode recordings, have a distinctive dendritic morphology with many apparently closed, or nearly closed, loops, giving the appearance of anastomosing dendrites (Fig. 13, Fig. 14). ON-OFF directionally selective ganglion cells are bistratified in the inner plexiform layer (Fig. 13), whereas ON-type directionally selective cells are monostratified, close to ganglion cell bodies (Fig. 14). None of these morphological features hints at the orientation of the null/preferred axis, however."
On "Color and Spectral Responses" "Ganglion cells respond to colored stimuli in one of two ways: color opponent responses and luminance responses. " "Goldfish are an animal model with color vision and have provided much information on the way in which vertebrate retinas process color." "In this animal, it was first realized that individual cones express only one each of three genomically available cone photopigment types (colloquially, red, green, or blue cones; technically, long (L), mid (M), or short (S) wavelength types). This supported at least part of Thomas Young's (48) 19th century "trichromatic" theory of color vision. "
"Tonic and Phasic Ganglion Cells of Primate Retina" "Tonic cells of the parvocellular pathway respond best to stimuli with high contrast and fine grain, whereas phasic cells respond to stimuli with very weak contrast, covering larger areas (58)." "Tonic cells respond to light stimuli in a steady maintained manner. Receptive field centers are extremely small, about 15 μm on the retinal surface (~4' of arc, about 10 cm at 100-m distance). Tonic cells are often called "midgets" because they probably represent recordings of midget ganglion cells described by Polyak (54) (see below). They occur with a density distribution across the retina comparable to, anatomically, midget cells (55)." "Phasic responses originate with morphologically larger ganglion cell types with fast optic nerve fiber conduction velocities (~4 m/s) (56)."
"X and Y Receptive Fields" "One of the assumptions of the DOG model (18) is that ganglion cells linearly add signals from both center and surround mechanisms for all points in space. This suggested that ganglion cells be tested for spatial linearity. Some cells (X-cells) pass the test, whereas others (Y-cells) do not, despite the fact that all receptive field data are well represented by the DOG model."
"In Fig. 11, impulse firing rate is shown for different positions (or spatial phase) of the sine wave stimulus with respect to the receptive field center. To change the spatial phase, the pattern is shifted right and left by incremental amounts. For the X-cell (Fig. 9A), when the pattern is positioned so that the transition from light to dark passes directly through the center of the field, a "null response" is produced. Introduction of the stimulus produces no effect on firing rate. The tendency of the bright bars to excite the cell is exactly compensated by the tendencies of the dark bars to reduce the firing rate."
"Ganglion Cells That Are Involved in the Circadian Clock" "Certain large-field, sparsely-branching ganglion cell types are known to project to the hypothalamic suprachiasmatic nucleus (SCN) in the brain (Fig. 49). The latter nucleus is the primary circadian oscillator in mammals and is essentially the biological clock, allowing resetting of bodily functions to local time. The ganglion cells involved in entraining the circadian clock project from the eye to the SCN via the retinohypothalamic tract (RHT) (Fig. 48). They apparently contain the neuropeptide pituitary adenylate cyclase activating polypeptide (PCAP), as well as glutamate."
a0007z.starburst a0309z.starburst
"Direction Selective Ganglion Cells (DSGC)" "Starburst Amacrine Cells (SAC)" "The ON-OFF DSGCs (ooDSGCs) each detect motion in one of four cardinal axes" "the ON DSGCs detect movement in the dorsal, ventral and nasal directions"
"By correlating physiological responses of each ooDSGC with structural evidence, from serial block-face electron microscopy, of putative synapses on these same cells, they showed that the SACs on the null side of ooDSGCs make more inhibitory synapses onto ooDSGC dendrites compared to SACs on the preferred side. In particular, individual SAC dendrites that are oriented antiparallel to the null direction of the ooDSGC contribute to the majority of these inhibitory synapses (fig.10). Hence, an increase in the number of SAC-ooDSGC inhibitory synapses, rather than the strength of individual ones, accounts for the direction-selective responses of ooDSGCs."
An increase in the number of inhibitory synapses, rather than the strength of individual synapses, accounts for direction selective responses (in the dendrites of individual neurons)
"Each SAC can wire to multiple ooDSGCs and contribute to their direction-selective responses. This is possible because each SAC dendrite is an independent direction-selective unit (Euler et al., 2002) (fig.11). When a stimulus is presented in a centrifugal direction, moving from its cell body out to the distal tips, the SAC dendrite inhibits the ooDSGC that it connects to. Therefore, each SAC dendrite responds best to motion in the opposite direction relative to that preferred by its postsynaptic ooDSGC (fig.12)."
The inhibitory inputs from the direction selective (SAC) Starburst Amacrine Cells cause pattern selectivity to happen in the more common On-Off Direction Selective Ganglion Cells (ooDGSCs)
By themselves the SAC cells are direction selective but through inhibitory signals they confer pattern selectivity to the more common ooDSGCs.
"On the other hand, SACs on the preferred side contribute relatively little inhibition."
"each SAC dendrite responds best to motion in the opposite direction relative to that preferred by its postsynaptic ooDSGC"
But the SAC is connected to the ooDSGC in a way that sets a pattern detection preference. (like is this cell going to activate when it sees pattern A (left to right movement), or Pattern B (dorsal to ventral movement (top to bottom))).
This paper is talking about Direction selectivity, but underlying that is the concept of pattern selectivity, and this is at the root of how memory works in the NAPOT theory. That long term memories are stored in the physical configuration of cells, with the synaptic connections & dendritic configurations, allowing a cell to respond preferentially to certain types of memories, so it activates when it sees what it has grown to look for, and it ignores (inhibits) patterns that do not match it's grown criteria. This is part of how long term learning grows in your mind. Cells learn to preferentially respond to certain patterns, and they learn to inhibit other patterns.
Inhibition then, (and we might apply this concept generally to inhibitory interneuron circuits) inhibition is like a second carving tool for the sculptor that is your brain rendering the mind, some of the patterns in your mind are rendered through excitatory circuits, but those patterns are improved & refined with inhibitory circuits. With inhibition what might otherwise be a rough & wacky hallucinatory experience becomes instead a refined reality, with more accurate pattern representation, and clearer thinking.
“GABAergic inhibition shapes behavior and neural dynamics in human visual working memory” https://academic.oup.com/cercor/advance-article/doi/10.1093/cercor/bhad522/7512628?login=false
"Choice selective inhibition drives stability and competition in decision circuits" https://www.nature.com/articles/s41467-023-35822-8
This concept also provides a clue to how Dendritic Inhibition might shape your active memory, by causing a cell to respond more or less to something in your receptive field.
What would it mean to apply the inhibitory concept to artificial neural networks such as Stable Diffusion?
In this next paper
"Directionally selective retinal ganglion cells suppress luminance responses during natural viewing" https://www.nature.com/articles/srep35708.pdf
It is shown that inhibition helps a direction selective cell to ignore stationary behavior. (or patterns that almost match, but do not actually match what the memory-prediction criteria that the cell is seeking to activate in your mind)
"Directionally selective (DS) ganglion cells of the retina report the direction of motion by spiking robustly to movement of an object in one direction (preferred) and sparsely to movement of the same object in the opposite (null) direction1–3. It is somewhat surprising therefore that such cells also spike robustly to changes in luminance without corresponding motion1–5 as responsiveness to two different visual features raises questions of how downstream circuits interpret the meaning of individual spikes, e.g. does a given spike convey information about motion or about luminance?"
"Responsiveness to multiple features by DS cells has been largely ignored in the feature detector literature, curious given that the identification of directional selectivity helped shape the feature detector hypothesis"
^ hmm hmm
"We find that responses to changes in luminance are strongly suppressed in favor of a response that reliably reports direction of motion."
^ Inhibition
"Coupled with the suppression of the response to luminance, this reduction of motion responses results in spike generation for only a single feature: motion of an object in the preferred direction. Responsiveness to only a single feature allows for unequivocal interpretation of transmitted spikes by downstream circuits and makes DS cells a true ‘feature detector.’"
"This finding is in contrast to the results of a recent study in which direction coding was not strongly affected by wide-field inhibition. Although the changes to the DS index in our study were modest, they arose consistently when the surround inhibition was blocked and thus highlight the importance of natural scenes for extracting certain characteristics from visual neurons"
"Synaptic Potentiation at Basal and Apical Dendrites of Hippocampal Pyramidal Neurons Involves Activation of a Distinct Set of Extracellular and Intracellular Molecular Cues" doi: 10.1093/cercor/bhx324
"In the central nervous system, several forms of experience-dependent plasticity, learning and memory require the activity dependent control of synaptic efficacy."
Synaptic Efficacy: The up & down regulation of individual synapses is part of how Neurons project or distribute sparsely coded patterns to the exit terminal array, but also how they backpropagate synaptic patterns backwards to the input terminal array.
NAPOT revision 2 is about attaching the concept of NAPOT to SDR at the synaptic scale. NAPOT Neural Array Projection Oscillation Tomography + SDR Sparse Distributed Representation with individual synaptic efficacy.
"Here we studied the functional and molecular aspects of hippocampal circuit plasticity by analyzing excitatory synapses at basal and apical dendrites of mouse hippocampal pyramidal cells (CA1 region) in acute brain slices."
"We demonstrate that synapse-specific molecular pathways allowing MMPs to rapidly upregulate function of NMDARs in stratum radiatum involve protease activated receptor 1 and intracellular kinases and GTPases activity. In contrast, MMP-independent scaling of synaptic strength in stratum oriens involves dopamine D1/D5 receptors and Src kinases. "
"Results of this study reveal that 2 neighboring synaptic systems differ significantly in extracellular and intracellular cascades that control synaptic gain and provide long-searched transduction pathways relevant for MMP-dependent synaptic plasticity."
Layer 5 & 6 Pyramidal Cells
"receive multiple inputs to 2 types of dendritic trees: basal and apical."
It's interesting that only pyramidal cells receive this description when back propagating action potentials happen to all neurons. Really the smaller neurons are smaller functional fractals of the larger neurons, so we ought to see back propagating action potentials from the postsynaptic membrane to the presynaptic membrane in all neurons, and glial cells, and also back propagating signals from the exit terminal branches (pre-synaptic) towards the soma. The same functions in larger pyramidal cells ought to appear in smaller neurons, because they are functional fractals of their larger cousins, meaning the same functions at a smaller scale.
back-propagating action potential (bpAP) "The back-propagating action potential (bpAP) is crucial for neuronal signal integration and synaptic plasticity in dendritic trees. Its properties (velocity and amplitude) can be affected by dendritic morphology. " "We found that the velocity of bpAPs was not uniform in a single dendrite, and the bpAP velocity differed among distinct dendrites of the same neuron." "The velocity of a bpAP was positively correlated with the diameter of the dendrite on which it propagated." "In addition, when bpAPs passed through a dendritic branch point, their velocity decreased significantly." "Similar to velocity, the amplitude of bpAPs was also positively correlated with dendritic diameter, and the attenuation patterns of bpAPs differed among different dendrites." "The amplitude of a bpAP may determine the strength of subsequent postsynaptic depolarization and affect the induction of plasticity. The velocity and frequency of bpAPs influence the timing of postsynaptic activity and the chance of potentiation. Therefore, the properties of bpAPs have a critical influence on the integration of synaptic input and the induction of synaptic plasticity" "The average velocity of bpAPs in granule cell dendrites is 150 μm/ms (150 micrometers per millisecond) which is slower than their velocity in pyramidal cell apical dendrites (500 μm/ms) and basal dendrites (200 μm/ms)" "The marked differences among bpAPs on the dendrites of different neuronal types inspired us to investigate the effect of morphology on bpAPs, with a focus on how the diameter and branch pattern influence bpAP velocity and amplitude."
"To overcome this limitation, recent studies have applied optical recording via genetically-encoded voltage indicators (GEVIs) or genetically-encoded Ca2+ indicators [21, 22]. Here we used a previously developed all-optical electrophysiological method using a GEVI [23] to record the membrane voltage in the dendrites of cultured hippocampal neurons."
"The results clearly showed that the bpAP propagation velocity in dendrites fuctuated (Fig. 1H)." "bpAP Properties Are Mainly Determined by Dendritic Tree Morphology" "After classifying the analyzed neurons into different types, we found that there was no significant difference in the average velocities of bpAPs in pyramidal neurons (190 μm/ms) and granule cells (202 μm/ms) (Fig. 2B)."
"We designed experiments to determine whether the volume of the cell body and the number of dendrites influences bpAP propagation."
"This finding indicated that the volume of the cell body and the number of primary dendrites have little influence on bpAP propagation."
"We found that 62% of dendrite pairs showed a positive correlation between the relative initial amplitude ratio and the diameter ratio, and this positive correlation became more significant as the diameter ratio increased (Fig. 4C). This finding indicated that bpAPs on dendrites with relatively large diameters are likely to also have relatively high amplitudes."
https://pubmed.ncbi.nlm.nih.gov/35984622/
"Synapse-Specific Regulation Revealed at Single Synapses Is Concealed When Recording Multiple Synapses"
"Synaptic transmission and its activity-dependent modulation, known as synaptic plasticity, are fundamental processes in nervous system function. Neurons may receive thousands of synaptic contacts, but synaptic regulation may occur only at individual or discrete subsets of synapses, which may have important consequences on the spatial extension of the modulation of synaptic information." "We have investigated how well-known synapse-specific short-term plasticity, where some synapses are regulated and others left unregulated, mediated by astrocytes and endocannabinoid (eCB) signaling can be assessed at different observational levels." "We show that eCB-induced depolarization-induced suppression of excitation (DSE) and astrocyte-mediated synaptic potentiation can be observed when monitoring single or few synapses, but are statistically concealed when recording the activity of a large number of synapses. These results indicate that the electrophysiological methodology is critical to properly assess synaptic changes occurring in subsets of synapses, and they suggest that relevant synapse-specific regulatory phenomena may be experimentally undetected but may have important implications in the spatial extension of synaptic plasticity phenomena."
"Synaptic efficacy refers to the strength of communication between neurons, and mainly depends on the probability and amount of neurotransmitter released from presynaptic neurons and the number of postsynaptic receptors activated"
"Therefore, three parameters define synaptic transmission properties in a single synapse: the probability of neurotransmitter release, the synaptic potency (i.e., the number of postsynaptic receptors activated), and the synaptic efficacy that results from the combination of them (del Castillo and Katz, 1954; Hessler et al., 1993; Dobrunz and Stevens, 1997; Atwood and Karunanithi, 2002)."
I think this next quote adds to the Cellular Oscillating Tomography theory even though it's about neurons, the principle ought to hold true for other types of cells. Meaning that the more receptors on a cell that are triggered by the same stimulus, the greater the cells response, because the cells morphological structure is inevitably going to sum up the aggregation of its receptor activations, and that ought to effect the cells reaction (a logical prediction based on an understanding of how physics works). "minimal stimulation activates a single fiber to observe the contained processing of a single synapse. Increasing fiber stimulation recruits synapse ensembles to study the aggregation of synapses in a cell, summing along dendritic arborization." See Figure 1 "FIGURE 1. Assessing synaptic transmission at different levels of analysis" https://www.frontiersin.org/articles/10.3389/fncel.2017.00367/full#:~:text=Synaptic%20efficacy%20refers%20to%20the,number%20of%20postsynaptic%20receptors%20activated.
"Synaptic Input and ACh Modulation Regulate Dendritic Ca2+ Spike Duration in Pyramidal Neurons, Directly Affecting Their Somatic Output" https://www.jneurosci.org/content/42/7/1184
"Nonlinear synaptic integration in dendrites is a fundamental aspect of neural computation. One such key mechanism is the Ca2+ spike at the apical tuft of pyramidal neurons. Characterized by a plateau potential sustained for tens of milliseconds, the Ca2+ spike amplifies excitatory input, facilitates somatic action potentials (APs), and promotes synaptic plasticity." "we explored the plateau and termination phases of the Ca2+ spike under input current perturbations, long-step current-injections, and variations in the dendritic high-voltage-activated Ca2+ conductance (that occur during cholinergic modulation)."
"We found that, surprisingly, timed excitatory input can shorten the Ca2+ spike duration while inhibitory input can either elongate or terminate it. A significant elongation also occurs when the high-voltage-activated Ca2+ channels (CaHVA) conductance is increased"
This finding is not surprising to me because there is an inverse relationship between magnitude & frequency. Excitatory input has higher frequency, so the magnitude (amplitude & duration) ought to decrease. It is natural for an excitatory (frequency) signal to have an inhibitory (duration) effect and vice versa, an inhibitory signal (frequency) might increase duration (part of magnitude) because of the inverse relationship.
"These Ca2+ spikes generate a huge depolarization in the dendrites, which lasts for tens of milliseconds and is accompanied by a significant influx of Ca2+ ions into the cell. The dendritic depolarization associated with the Ca2+ spike often affects the somatic region and leads to a short burst of axonal APs (Leleo and Segev, 2021). Functionally, Ca2+ spikes were shown to play a central role in the transmission of higher-level top-down signals, the coupling between the soma and the apical tuft, and the modulation of synaptic plasticity "
"Because of the relatively long duration of the dendritic Ca2+ spike, various manipulations and synaptic input perturbations may alter its stereotypical voltage waveform. Thus, each of these manipulations can directly affect local plasticity, the communication between the soma and the nexus, and the overall output of the cell. "
See Figure 1 "The Ca2+ spike can be divided into three phases: (P1) initiation, (P2) plateau potential, and (P3) termination. D, Quantification of the variability in trace shape for each phase in the voltage waveform, in response to a range of current stimuli." ^ Reference the link above
In a nutshell: individual synaptic back propagation, from another cells dendrite, or from this cells dendrite, is going to affect the Ca2+ Calcium back propagation waveform, which will affect whether the Basal & Apical Dendrites will jointly produce a coincidence detection leading to burst firing from the Soma, shooting a memory prediction back out the Apical Dendrite to other neurons, with individual messages connected to each Apical Synaptic branch, allowing for a sparsely coded memory to be upload from this one cell to the neural array defined by it's exit terminal. Integrating a memory-prediction rendered pattern into your phenomenologically conscious mind.
Re-Read this paper 6 times
Neuroelectric Tuning of Cortical Oscillations by Apical Dendrites in Loop Circuits https://www.frontiersin.org/articles/10.3389/fnsys.2017.00037/full
"Many layer 5 and 6 pyramidal neurons are connected to thalamic neurons in loop circuits"
"It is proposed that a major function of the apical dendrite is to produce sustained oscillations at a specific frequency that can serve as a common timing unit for the processing of information in circuits connected to that apical dendrite."
"Synchronous pulse outputs from the circuit loops containing apical dendrites can tune subthreshold membrane oscillations of neurons they contact. When the pulse outputs are finely tuned, they function as a local “clock,” which enables the contacted neurons to synchronously communicate with each other. Thus, a shared tuning frequency can select neurons for membership in a circuit. Unlike layer 6 apical dendrites, layer 5 apical dendrites can produce burst firing in many of their neurons, which increases the amplitude of signals in the neurons they contact. This difference in amplitude of signals serves as the basis of selecting a sub-circuit for specialized processing (e.g., sustained attention) within the typically larger layer 6-based circuit. After examining the sustaining of oscillations in loop circuits and the processing of spikes in network circuits, we propose that cortical functioning can be globally viewed as two systems: a loop system and a network system. The loop system oscillations influence the network system’s timing and amplitude of pulse signals, both of which can select circuits that are momentarily dominant in cortical activity."
"The prevailing view of the main function of the apical and basal dendrites of a pyramidal neuron is the integration of incoming electric pulses at the tens of thousands of synapses which dot the dendrite surface (e.g., Häusser et al., 2000; Spruston, 2008) The result of this integration is a train of output pulses in the single axon which exits at the base of the soma."
It exits the base of the Soma to the Apical Dendrite right???
"Here, we propose that a major function of the long apical dendrite in pyramidal neurons is the production of a stable oscillation at a specific frequency."
"When apical dendrites in corticothalamic loops oscillate, their oscillations are copied to networks of pyramidal neurons, and all network pyramidal neurons that are contacted will oscillate at a common carrier frequency into which messages of temporally coded spikes may be inserted."
"Blocked" Reference 1
"Only neurons whose membranes oscillate at the specific carrier frequency will accept the spike signals: spike signals carried on other frequencies will be blocked."
This confirms for me the concept I have that the synchronously oscillating group is learning together, and that oscillatory synching between different oscillating groups is how the brain relays information to other parts of the brain, the concept that I said was similar to Neuropype technology (by Intheon CEO Tim Mullen) relating unstructured heart data to unstructured eeg data, or any kind of sensor stream data with a time code. I think the brain works the same way and this note seems like evidence for the idea.
One of my general arguments seems to be that the function of synchronous oscillations can be compared to the function of noise in a Stable Diffusion neural network. It's allowing the neural network of the brain to absorb the high phasic spikes across the oscillating cell assembly, so each cell is learning it's own variation on the received pattern, each cell is capable of participating in a learned sequence recall that involves many cells firing across many layers of cortex inside a cortical column, or across cortical columns in the regional networks that span multiple cortical columns & the thalamus.
"The corticothalamic loops operate as a clock in a computer, which assures that electric signals in different locations within a connected circuit of neurons change at the same time. In the present theory, different bundles of apical dendrites can oscillate at different frequencies, and therefore their connected circuits of neurons can run on different clocks."
The main concept of NAPOT applied to phase changes between large oscillating groups basically amounts to changes in the clock speed of one corticothalamic loop, that is absorbed by the rest of the cortical thalamic loops, via the physics of oscillation, and those differences in clock speed are the phase changes created by the aggregates of phase wave spiking, and this is how signaling works at the macroscopic scale in the brain between large clusters of oscillating cell assemblies.
The medial or mesoscopic brain activity is about how layers of neural circuits within cortical columns & within the hippocampus talk to each other.
"The axons of pyramidal neurons of layers 2 and 3 form an information bridge from one minicolumn to other minicolumns at locations both near and far. Axons of layer 5 pyramids contact minicolumns near and far, so that oscillations of the long apical dendrites positioned vertically at the center of a minicolumn can be synchronized with oscillations of neurons throughout a broad network of neurons. When membranes of network neurons oscillate in synchrony the wave peaks of these oscillations are separated by a constant time interval; so it can be said that these neurons bind together and “talk to each other” (Fries, 2005, 2015; Womelsdorf et al., 2007)."
"To summarize this section"
1. The oscillations of layer 5 and 6 apical dendrites drive the output pulses of their pyramidal axons, which are copied to the thalamus and sent back to the same apical dendrite, forming a loop circuit."
2. "The thalamic axon also sends these pulses to other neurons to which it is connected, some located nearby, some located remotely. In this manner, oscillations of the same frequency as those in the initiating apical dendrites can be spread across the cortex."
"The modeling of oscillations at any given location along the apical dendrite is represented theoretically by a distribution of resonance frequencies, which is visually depicted as a resonance profile. Two features of the profile are evident and of main interest: its peak and its spread. The peak frequency of the resonance profile is customarily called the resonant frequency. As it is most generally used, the term resonance denotes the ability of a system to oscillate most strongly at a particular frequency. It may be noted that the existence of resonance in this biophysical model does not require inductance, in contrast with most classical electrical circuitry that is tuned to a specific frequency."
"Both the peak resonance frequency and the progressive narrowing process are assumed to be influenced by many features of the membrane, but chief among these features is the inward flow of positive sodium ions by spines, the consequent outward flow of positive potassium ions around the spines, the passive outflow of potassium ions by the leak current, and to a lesser extent the activities of many other types of channels in the dendritic membrane. The initial surge of charge consists of action potentials which are delivered by a thalamic axon at the top of the apical dendrite"
"Thus, the present model predicts that the rate of outward potassium flow through the apical dendrite membrane determines the peak frequency of the dendrite’s oscillations."
^ This quote above is consistent with NAPOT theory about the role of potassium changing the magnitude & its inverse which is the frequency (because magnitude & frequency have an inverse relationship)
"This variable level of voltage amplitude of a current surge in the apical dendrite contrasts with the discrete, all-or none voltage amplitude of an action potential spike propagating along an axon of the neuron."
The physics here force a change in the APD Action Potential Duration. If you have fixed amplitudes (the all or none action potential) then excess voltage beyond the triggering threshold is going to change the wavelength (and frequency), leaving Calcium channels open longer, changing the message that is sent from that neuron into the network. A change in the timing of the oscillating clock relative to the other oscillating clocks that it was in sync with. That phase wave time change is information. The smaller oscillation now exerts a mutual force or drag on the larger group oscillation.
"As current surges flow down the apical dendrite their oscillation frequency is assumed to change slightly when local potassium channels change their rate of releasing potassium ions from the dendrite."
"Low spine voltage produces low outward potassium flow and low values of oscillation frequency, and high spine voltage produces high potassium flow and high values of oscillation frequency. Thus, The present model predicts that the rate of outward potassium flow through the apical dendrite membrane determines the peak frequency of the dendrite’s oscillations."
Medical Imaging Tech Sidebar
"Measurements of these oscillations can be obtained from the local electric field potentials that they produce; and as the field potentials of clusters (bundles) of thousands of apical dendrites radiate to the scalp their combined voltage can be measured as EEGs."
ie EEG Electroencephalogram electrodes on your scalp are detecting voltages from the electric dipoles (that represent the combined voltages of thousands of individual apical dendrites.)
"Peak Frequency as a Function of Rate of Ion Outflow Four equally spaced frequencies in the 0–100 Hz range were selected, 20, 40, 60, and 80 Hz, and the outward flow of potassium ions, gr, was calculated for each frequency using Equation [1] of Kasevich and LaBerge (2011), which is based on a single compartment. Iterative calculations converged to a value of gr that maximized the energy transfer from one compartment to the next compartment. Appropriate geometric parameters of the underlying leaky cable theory and electrical parameters of the membrane were obtained from the relevant literature and entered into Equation [12] of the Kasevich and LaBerge (2011) study."
"The measure of energy transfer was transfer impedance. The relationship between the obtained gr conductance and each of the four frequency points is graphed in Figure 6."
See figure 6 in the above link (in the title) "FIGURE 6. Simulated relationship of profile peak frequency as a function of the rate of outward conductance of potassium ions. The obtained outward potassium conductances were 105, 310, 627, and 1056 nS for the curves with peaks at 20, 40, 60, and 80 Hz, respectively. The resonant frequency in radians per second is approximately the radial conductance divided by the radial capacitance. For a constant radial capacitance the peak frequency is directly proportional to the potassium ion leak conductance. [Adapted from Kasevich and LaBerge (2011)]."
"FIGURE 6 | Simulated relationship of profile peak frequency as a function of the rate of outward conductance of potassium ions. The obtained outward potassium conductances were 105, 310, 627, and 1056 nS for the curves with peaks at 20, 40, 60, and 80 Hz, respectively. The resonant frequency in radians per second is approximately the radial conductance divided by the radial capacitance. For a constant radial capacitance the peak frequency is directly proportional to the potassium ion leak conductance."
"The peak frequency is directly proportional to the potassium ion leak conductance."
This means that the amount of potassium ions in the neuron when the action potential fires determines the frequency & magnitude of the phase wave that bursts from the soma and travels along the axon triggering vesicle release
(note the frequency of muscle movement 20–300 Hz)
"If we assume that neural noise at each synapse, on average, adds a constant amount of impedance (opposition) for each frequency across the 1–100 Hz range, then each synapse will dampen or flatten the shape of the oscillation frequency profile."
An observation consistent with the NAPOT concept that oscillatory synchrony is noising (the opposite of denoising) and absorbing high phasic bursts through oscillatory dissipation of the high phasic (action potential or action potential burst sequence) phase wave differential.
"While every cycle of the corticothalamic circuit adds noise to flatten the shape of the profile curve, the next cascade through the apical dendrite compartments narrows the shape of the profile."
An observation consistent with the NAPOT theory regarding modeling each high phasic action potential as a tensor in a Taylor Series, that can learn a curve or a shape as signals are passed through successive neural arrays.
"Therefore, increasing the number of compartments in the apical dendrite provides a means for effectively offsetting the synaptic noise in the corticothalamic circuit.
"Hence, longer apical dendrites (e.g., in primates) should produce more narrow frequency profiles, while very short apical dendrites (e.g., in mice) should show a limit on the narrowness of the e profile. For a comparison of apical dendrite lengths of layer 5 and layer 2/3 pyramidal neurons across five mammalian species see Figure 1 in LaBerge (2005)."
^ Relevant to the point I made in note a0008z and in this note about what Neurogrid misses by virtualizing the dendritic branches, search this note for lengthreference1
TWO TUNING NEURONS The Pyramidal Neuron of Layer 6
"The pyramidal neurons of layer 6 are widely diversified in area V1 into at least eight different types, which are defined by their patterns of dendritic inputs and axon outputs (Briggs, 2010). Sherman and Guillery (1998) described particular layer 6 pyramidal neurons whose axons appear to modulate, as opposed to driving, the processing of neurons they contact (Figures 9, 10). Examples of driver axons are the output fibers arising from thalamic neurons which directly contact stellate neurons of layer 4, and whose axons presumably produce spikes. Examples of modulatory axons arise from layer 6 pyramidal neurons and contact apical dendrites of layers 2/3, 5, and 6 pyramidal neurons as well as the stellate neurons of layer 4 (Figure 9)."
Drive vs Modulate, the driving neuron passes the signal, it doesn't change it, it's not adding information. Driver neurons are being told where to go. Modulating neurons are changing the signal as it passes. So this paper might seem to suggest that Layer 6 Pyramidal Neurons tuning as a passive activity, and that Layer 5 is doing the real work of driving change, the opposite is true. In the neurophysics context driving is more passive, and tuning involves information generation.
"Thalamic neurons that project to the cortex are of two main types, called “core” neurons and “matrix” neurons, which are distinguished by their immunoreactivity for the calbindin calcium-binding protein and the parvalbumin calcium-binding protein, respectively (Jones, 1998, 2007)"
"The core thalamic neurons project to middle layers of the cortex, (Figure 9) (...) and the matrix thalamic neurons project to superficial layers 2 and 3 of the cortex (Figures 13, 14)"
"The findings of Sherman and Guillery (1998) support the view that the direct connection between the thalamus and a stellate neuron carries a series of spikes that codes the information arising from sensory receptors (e.g., in the retina), and in higher-order thalamic neurons codes the information arising from other cortical areas (Mitchell et al., 2014). In In the context of the present theory, the connection between the layer 6 tuning-neuron and a stellate neuron provides precisely timed intervals which adjust the series of spikes into a stable temporal code."
"In this way, the information in the sensory receptor is transmitted to cortical networks in the form of a precise temporal code."
"Blocked" Reference 2
"Because the membranes of neurons that form a specific circuit in the networks of information processing oscillate at the same frequency, the neurons can receive coded pulse trains from each other. Thus, the frequency of oscillation acts like a frequency channel in two-way radio communications. The preference of these neurons for one particular frequency implies that spike communications coded at a different frequency will be blocked at the membrane."
"Blocked" Reference 3
"One way to enable one neuron to dominate another neuron has already been described here for layer 6 axons. When two layer 6 axons compete for dominance at a dendritic or somatic membrane, the one whose spike frequency matches that of the membrane’s ongoing oscillation succeeds in delivering its series of spikes; if the other axon responds at some other spike frequency its input will be blocked. Conventional metaphors for This blocking type of selective mechanism is a filter or gate."
(Blocking or Gating is usually caused by the inhibition of a receptor right?)
"Blocked" Reference 4
"When apical dendrites in corticothalamic loops oscillate, their oscillations are copied to networks of pyramidal neurons, and all network pyramidal neurons that are contacted will oscillate at a common carrier frequency into which messages of temporally coded spikes may be inserted. Only neurons whose membranes oscillate at the specific carrier frequency will accept the spike signals: spike signals carried on other frequencies will be blocked."
Sitting patterns
NAPOT Revision 4 is essentially that patterns sit in our minds as active synapses between neurons that are firing tonically at the same frequency, and these computational mental renderings are also defined by the inhibited synapses between neurons that are not firing at the same resonant frequency.
In NAPOT Revision 2 I theorized that individually upregulated & downregulated synaptic channels represented sparse & distributed representations that became printed to the network with a burst firing or high phasic somatic action potential.
But those synaptic patterns also oscillate with regular tonic action potentials, so they are printed regularly, to the mind, to become part of the rendered pattern of the mind.
I learned about the research on the Clusteron that I wrote about in note a0138z https://github.com/v5ma/selfawarenetworks/blob/main/a0138z.md
That note dialed into how different thresholds for calcium can yield polar opposite cell behaviors, in their model if the synapse's medium threshold is reached the calcium triggers the LTD process (the receptors face removal via endocytosis (falling into the cell) or exocytosis(falling out of the cell)), (those lost receptors may move to a new location on the dendrite), and if the synapses high threshold is reached the calcium triggers the LTP process (new protein synthesis bolsters the placement of receptors or causes more receptor growth in that area).
"How the Brain Focuses on What’s in Mind" https://neurosciencenews.com/pfc-varability-focus-21363/ Reduced variability of bursting activity during working memory https://www.nature.com/articles/s41598-022-18577-y
Which I tagged telescoping mind dials in on the fact that activated cell firing not only oscillates but via inhibiting it's exit terminal it sets the oscillatory timing of a bunch of other cells, magnifying a memory to it's network, but also making the memory persist without new activity needed.
A memory once initiated with a phasic action potential continues to oscillate with tonic oscillations, like a new expectation has been loaded into working memory. (In theory a negative action potential (slow potential or inhibitory action potential) should kill it (remove it from active expectation or working memory), and a high phasic (bursting) should amplify it into your focus perhaps taking over enough of a dendritic branch to cause a neuron representing another learned pattern to fire in phasic or bursting mode.)
The delta of potassium ions in the neuron when the action potential fires determines the frequency that neuron will create: 20 (Beta), 40 (Gamma), 60 (Gamma), and 80 (high Gamma) Hz"
"Only neurons whose membranes oscillate at the specific carrier frequency will accept the spike signals: spike signals carried on other frequencies will be blocked."
In NAPOT 1 I covered the concept that a dendrite & its spines (receptors) can represent a memory that is defined by its synaptic connections and its morphology which determine what patterns the neuron is receptive to and what it ignores.
In NAPOT 2 I explored how inhibited & excited synapses might represent sparse distributed memory-predictions or SDR Sparse Distributed Representations, that might be stored in a synaptic configuration in the inceptive field of the exit terminal or apical dendritic branches.
This was a good talk on 𝗦𝗽𝗶𝗸𝗲𝘀, 𝗟𝗙𝗣𝘀, 𝗮𝗻𝗱 𝗪𝗮𝘃𝗲𝘀 https://youtu.be/qftGEaWKTfI I enjoyed this but at the same time it felt disorganized & I felt that they failed to really discuss the stated topic of the discussion, which was whether or not brainwaves have a non-incidental function in cognition. In the talk it was speculated by one speaker that the signifance of emphaptic coupling could be trivial or incidental or not worthy of being the focus of his research but I think the research shows that Emphaptic coupling is essential to brain & heart activity, and we could not function without it.
In NAPOT 3 I explored the SOMA scale anti-spike, the slow potential, slow wave potential, slow cortical potential, the DC potential, or the inhibitory delta frequency spike waveform from the SOMA might convey information, and separately how the extracellular potassium gradient contributing to the signals we see in local field potential measurements, eeg, and brain waves that carry the aggregrate representatives of many phase changes (the signature of burstlets) from the extracellular ionic & transmitter gradients from neurons & glia.
Local field potential https://en.wikipedia.org/wiki/Local_field_potential#:~:text=LFP%20are%20typically%20recorded%20with,in%20vitro%20brain%20thin%20slice.
In otherwords, in NAPOT 3, I'm speculating that primarily the extracellular potassium gradient, secondarily the extracellular calcium gradient, and to a lessor extent changes in the sodium & chloride gradients resulting from spiking activity give rise to local field potentials, which gives rise to Ephaphtic coupling events.
Voltage gated channels as well as Potassium & Calcium activated channels are going to be affected, leading to the inhibition or excitation of individual synapses via Ephaptic coupling. In this way brainwaves might be felt by the brain (perhaps as feelings or emotions which I described in NAPOT 5 as being a match for tonic high magnitude low frequency brainwaves).
Brainwave Oscillations, it has been argued by Buzsaki (Rhythms of the Brain, 2006), and others, help to bind & synchronize brainwave activity into powerbands with the names delta, theta, alpha, beta, gamma, high gamma etc... these oscillating powerbands serve as binding attractors that absorb discordent oscillating states (like a de-resonating attractor that essentially dissipates the resonating oscillations of burst firing events including sharp wave ripples into the higher magnitude & lower frequency powerband) or otherwise the powerband oscillations exert an effect that is non-synchronous (describing something like a chaotic attractor, or possibly something like a splay state where a non-synchronous interaction is perpetuated over time).
Ephaptic coupling has also been documented as having a role in synchronization and timing with the mechanism being oscillatory physics as described by Steven Strogatz in his book Sync for example. I'm arguing that the synchronization via ephaptic coupling is analogous to the signals sent between oscillating clocks or between fireflies that eventually causes synchronization.
Ephaptic coupling via Wikipedia: "For example, the currents that caused the depolarization (excitation) of the active nerve caused a corresponding hyperpolarization (depression) of the adjacent resting fiber. Similarly, the currents that caused repolarization of the active nerve caused slight depolarization in the resting fiber. Katz and Schmitt also observed that stimulation of both nerves could cause interference effects. Simultaneous action potential firing caused interference and resulted in decreased conduction velocity, while slightly offset stimulation resulted in synchronization of the two impulses."
"conditions can be manipulated in such a way that the action potential from one neuron can be spread to a neighboring neuron. This was accomplished in one study in two experimental conditions: increased calcium concentrations, which lowered the threshold potential, or by submerging the axons in mineral oil, which increased resistance. While these manipulations do not reflect normal conditions, they do highlight the mechanisms behind ephaptic excitation"
"Ephaptic coupling has also been found to play an important role in inhibition of neighboring neurons. Depending on the location and identity of the neurons, various mechanisms have been found to underlie ephaptic inhibition. In one study, newly excited neighboring neurons interfered with already sustained currents, thus lowering the extracellular potential and depolarizing the neuron in relation to its surrounding environment, effectively inhibiting the action potential's propagation."
" In the simpler case of adjacent fibers that experience simultaneous stimulation the impulse is slowed because both fibers are limited to exchange ions solely with the interstitial fluid (increasing the resistance of the nerve). Slightly offset impulses (conduction velocities differing by less than 10%) are able to exchange ions constructively and the action potentials propagate slightly out of phase at the same velocity. More recent research, however, has focused on the more general case of electric fields that affect a variety of neurons. It has been observed that local field potentials in cortical neurons can serve to synchronize neuronal activity."
"it is hypothesized that neurons are ephaptically coupled to the frequencies of the local field potential. This coupling may effectively synchronize neurons into periods of enhanced excitability (or depression) and allow for specific patterns of action potential timing (often referred to as spike timing). This effect has been demonstrated and modeled in a variety of cases.[10][11]"
"Ephaptic interactions among cardiac cells help fill in the gaps that electrical synapses alone cannot account for. There are also a number of mathematical models that more recently incorporate ephaptic coupling into predictions about electrical conductance in the heart."
"The inhibition due to ephaptic coupling would help account for the integration of signals that gives rise to more nuanced perception of smells"
Helmut Schmidt ,Gerald Hahn,Gustavo Deco, "We demonstrate that, although the extracellular potentials generated by single spikes are of the order of microvolts, the collective extracellular potential generated by spike volleys can reach several millivolts. As a consequence, the resulting depolarisation of the axonal membranes increases the velocity of spikes, and therefore reduces axonal delays between brain areas. Driving a neural mass model with such spike volleys, we further demonstrate that only ephaptic coupling can explain the reduction of stimulus latencies with increased stimulus intensities, as observed in many psychological experiments."
"The key finding of our study is that spike volleys generate EPs with sufficiently large amplitudes to modulate axonal delays. Specifically, the mean delay of a spike volley decreases as the number of spikes in the spike volley is increased. Therefore, our results suggest that varying the amplitude of a neuronal signal can adjust its delay. Using a neural mass model, we have demonstrated that the decrease of axonal delays translates into the decrease of response latencies as the stimulus intensity is increased."
"In the presence of ephaptic coupling, the spike volley slowed down. This is in line with previous numerical studies which investigated ephaptic coupling effects between a small number of identical axons. There, ephaptic coupling led to a synchronisation of the spikes within a volley, and a concurrent deceleration of the spikes."
"the EP at a node of Ranvier can reach several hundreds of microvolts." "As oligodendrocytes can myelinate multiple axons [41, 42], it is conceivable that neighbouring axons show some degree of alignments, in which case it would be possible to observe ephaptic coupling effects in much smaller fibre bundles, provided the spike volleys are sufficiently synchronised."
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1007858
NAPOT 4 Converges 2 & 3 with the unification of two concepts: Synaptic Inhibition that is based on Somatic Frequency to create stable short term patterns in the mind.
NAPOT 4 helps connect together the different types of signals at the synapse scale exploring how they intersect with the different tonic frequencies that neurons can oscillate with. The Neuron can subselect active patterns, as active synapses, with its potassium gradient, so that it is only receiving synaptic patterns in a certain frequency range such as 20hz, 40hz, 60hz, or 80hz. Which means that synapses that are connected to neurons that are not in a matching frequency become inactive, blocked, filtered, gated, or inhibited.
So in NAPOT 1 we already covered how the dendritic morphology & the synaptic growth define a memory. In NAPOT 4 we explore how the frequency of the neuron, set by the soma burst, determined by the potassium quantity, selects which synapses are inhibited at a high level, blocking communication with neurons that are out of sync.
So essentially NAPOT 4 defines how persistent long lasting memory-predictions & internal representations can be redefined in real time, faster than it takes for receptors to grow or decay, because we are talking about the inhibition or excitation of individual receptors.
"How the Brain Focuses on What’s in Mind" (search for news article by this name)
Reduced variability of bursting activity during working memory https://www.nature.com/articles/s41598-022-18577-y
What I said previously in this article, linked above, highlights the fact that when cells fire they oscillate repeatedly. They are setting a pattern in motion. That pattern sits in tonic oscillation until it is changed. So patterns sit in our mind, like the led lights staying on, until they are changed, and they sit in the activated synapses, and these patterns are conversely also etched by the inhibited neurons between cells that are not oscillating together at the same resonant frequency.
The idea is that when you are focused on something, at a small scale, a neuron fired, it magnified it's message to it's network, it set the timing of it's inceptive field, so that those neurons that were tuned in or receptive to it's inceptive field (not blocking or inhibiting communication because they were in a different frequency) are going to become inhibited, so attention (reference: attention schema theory) becomes focused on the neuron that fired, and it's signal is amplified via the inhibition of nearby neurons, and it continues to make sharp wave ripples across the brain to neurons that are receptive to its inception.
So the idea from NAPOT 2 is that tonic action potentials are printing the synaptic configuration. Sending a sparse distributed message, but specifically the phase wave variance releases a quantity of vesicles per synapse between 0 1 2 3 with the quantity determined by APD action potential duration.
In NAPOT 4 I am exploring the idea that the phase change of Neuron C, changes which other neurons (D, E, F) that Neuron C is exchanging messages with, because that neuron's phase pattern is now firing at 20hz, 40hz, 60hz, or 80hz (when it's not inhibited).
So let's say that Neuron C at (20hz) isn't talking to D (at 40hz), E (at 60hz), or F (at 80hz), then it has an action potential spike with a major change in potassium levels resulting in a shift in its phase (frequency) from 20hz to 60hz. The idea is that now the synapses between Neuron C & Neuron E are uninhibited, communication channels are open between, they neurons are oscillating at the same frequency, and the synapses they share are upregulated, and so they share information (via neurotransmitters), it's bidirectional synaptic communication, and they act as a unified or entified sensor & transmitter system. They become a cell assembly that has wired together in the Hebbian sense.
I am exploring the idea that the postsynaptic membrane on a neuron is going to change its status based on the signal it receives, which is based on the phase pattern of its axon or neuron body.
So this is like if Neuron A shifts into a 40hz oscillation, and it sends a signal to a matching 40hz oscillating neuron, their synaptic channels will remain open to one another, it's sort of like "we both see this pattern" but then the message sent by 0, 1, 2, 3 vesicles per synapse is going to dial in the computations for the next computational pattern rendering.
It's like look around you, what you are seeing is not reality, it's a replication of reality, a computed rendered in the volumetric computer screen of active synaptic connections.
a0138z.clusteron reference the Numenta Video Clusteron "Plasticity and Learning Algorithms in Models of the Single Neuron"
With NAPOT 4 The discussion is about the blocking of individual synapses, based on the oscillation frequency set by the whole neuron.
The questions to keep researching to validate this are:
If we are looking at two neurons that are linked: A is the presynaptic, B is a postsynaptic neuron, can either neuron A or B initiate the blocking of receptors, by inhibiting one or both of the channels between A & B?
If A inhibits the synapse shared with B do both synapses become inhibited together?
Do both synapses become uninhibited together?
Are there situations where A's synapse can inhibit B's synapse but not its own, can B backpropagate to inhibit A's synapse but not its own.
Can A uninhibit its own synapse and then uninhibit or inhibit B's receptor?
Can B uninhibit its own synapse and then uninhibit or inhibit A's receptor?
What happens to the pre-synapse, and neural synchrony when drugs become receptor agonists with SSRI, DMT, LSD, Caffeine, and other mind altering food & drugs?
If A is oscillating at 40hz and B is oscillating at 20hz are all the synapses between them inhibited?
Does research exist that confirms that this process allows active internal representations to stay active in memory, like LED lights that stay on (as if with each soma tonic action potential oscillation the led of active synapses is imbued into the phenomenologically conscious mind) like patterns sitting in your mental workspace.
It's like the phase differential of redness is some configuration of upregulated synapses shared by neurons oscillating at the same frequency, defined also by the synapses that are not active, like the pixels representing black color on your tv monitor.
In the NAPOT 4 picture we are focusing on two types of neuropaths: One defined by grown biological connections (LTP synaptic growth/LTD synaptic atrophy) & Dendritic Morphology, and the other neuropath is defined by shorter term active & inactive synapses, that will change more rapidly with somatic phase changes. Which is useful for shorter term memory that needs to change on a faster time scale. The latter faster changing neuropath is essentially an electrically active subset of the longer term neuropath, and can change on a time scale ranging from milliseconds (NMDA & AMPA receptors) to seconds (BTSP Behavior Timescale Synaptic Plasticity)
b0232y.btsp https://github.com/v5ma/selfawarenetworks/blob/main/b0232y.md
These synaptic changes basically shape the wave path of phase wave signals as they travel through the brain, it shapes the path of voltage cell by cell, and going back again to the concept of phase wave changes being like tensors in a Taylor Series evoking shapes that your neural network can trace. Your internal representations that exist as phase changes across configurations of cells actively linked together, or not actively linked together.
a0149z https://github.com/v5ma/selfawarenetworks/blob/main/a0149z.md
In this thought experiment, networks of neurons with their connections create structures defined by active & inactive synapses that other neurons can recognize, as the brain's tonic oscillations dissipate signal energy valence the impressions of synaptically modeled information is distributed across the brain's network.
Internal representations defined by inhibited synapses (disconnected neurons) and excited synapses (neurons with the same frequency vibration) and these internal representations are also defined by the morphology of the dendrite, and the synaptic connections, and the parts of the dendrite that are active (because the voltage in that part of the dendrite opens up it's receptivity to associated patterns, because the voltage spill over from heterosynaptic plasticity increases the chances of synaptic depolarization from a lower frequency synaptic transmission)
The mirror image of the somatic phase change may also be released as a brainwave via the extracellular potassium gradient when the neuron acts as an electrical sink, grounding its extra energy to the brain as a whole.
"The preference of these neurons for one particular frequency implies that spike communications coded at a different frequency will be blocked at the membrane."
"(...)plausible that the frequency of coded pulses can serve as the basis for selecting particular cortical circuits, and for selecting the corresponding cognitive function(s) that they provide."
"In short, the layer 6 apical dendrites of a column cluster of minicolumns may serve to group neurons into functional circuits."
"Cognitive processing of a visual scene usually requires that a part of the scene is selected for special processing by an appropriate circuit or a part of a circuit. This more restrictive selective activity is commonly called attention. The particular form of attention being treated here is called sustained attention,"
"When sustained activity functions as sustained attention, additional neural mechanisms appear to be involved."
"an apical dendrite in a layer 6 pyramid loop communicates its particular oscillation frequency to a group of cortical neurons, which are then able to interact with each other by signals traveling on that particular “carrier frequency.”"
"when the membranes of dendrites and somas of two or more neurons oscillate at a common frequency (but at subthreshold voltages), only incoming axon spikes at that frequency will be accepted by the neurons."
To execute a task
"a specific task, e.g., to a search task, a response planning task, or a perceptual preparation task. "
Your 6th layer pyramidal neuron needs to create a cell assembly, imagine neural circuits as lego blocks that your brain has to stack to build a program to accomplish a task, repurposing other neurons & columns for the task at hand. "let's search for something" So it has to compete with other pyramidal circuits that are doing other tasks. "I'm eating"
In this paper the 6th pyramidal has to compete to dominate other neurons, to get the neural circuits it needs to complete a task.
"When two layer 6 axons compete for dominance at a dendritic or somatic membrane, the one whose spike frequency matches that of the membrane’s ongoing oscillation succeeds in delivering its series of spikes; if the other axon responds at some other spike frequency its input will be blocked"
A neuron in layer 5 will join your task, because it was oscillating at the same general frequency. I was feeling it. Same vibration.
"Another way to enable a particular axon to dominate other axons that synapse on a common dendritic segment is by shifting to the mode of bursting. When the signal amplitude of the target axon is sufficiently greater than that of other axons in the synaptic vicinity the signal inputs from other axons will be blocked"
Bursting mode essentially changes which neuron can talk to which other neuron, for the duration of that bursting.
"In electrical physics an appropriate metaphor is gain (or voltage) control, corresponding to the term enhancement in psychology. Another cognitive metaphor for this amplifying type of selection is a spotlight."
"FIGURE 11 | Proposed graphic descriptions of regular spiking and intrinsic bursting neurons. Layer 6 pyramidal neurons produce regular spiking, which is represented by a narrow carrot-shaped icon (intended to resemble the electric field of the active apical dendrite). The tapering carrot shape reflects the diminishing electrical effect of the persistent leak current, by which positive ions of potassium flow out of the dendrite. Layer 5 pyramidal neurons produce regular spiking, but also intrinsic bursting, which is a series of brief groups of very rapid spikes. As the number of spikes in a burst increases, the voltage of the burst pulse increases. To represent these differences in burst intensity the carrot-shaped icon varies in width."
The Thalamic Matrix Neurons create loops that connect cortical columns loops together (the neighborhood feedback chat)
"Axons of thalamic matrix neurons tend to contact pyramidal neurons within the minicolumn that originally sent axons to the thalamic matrix neurons"
"but they also contact pyramidal neurons in minicolumns that are located at adjacent cortical areas"
"FIGURE 12: The figure shows the carrot-shaped graphics of apical dendrite activity, which depicts the shape of the electric field surrounding the apical dendrite."
"Axons of core thalamic neurons tend to contact neurons located within the cortical column of origin"
The Core Thalamic neurons create loops with the cortical column of origin. (The mirror feedback)
Because of the thalamic connection between layer 5 pyramidal cells and layer 2/3 pyramidal cells (the connection between the distal ends of the apical dendrites)
"the apical dendrites of layer 2/3 pyramids will engage in oscillating activity that matches the oscillation frequency of the layer 5 pyramids. "
"When two layer 2/3 pyramidal neurons, separated by small or large cortical distances, oscillate in synchrony, they can effectively communicate with others."
"With the apical dendrites of layer 2/3 pyramids oscillating at the same frequency as the layer 5 pyramids, and with the layer 6 pyramids setting the oscillation frequency of the layer 5 pyramids, all of the pyramidal apical dendrites of the minicolumn apparently oscillate in synchrony."
This appears to be confirmation of the idea that a cortical column (in this case a mini column) represents an oscillating cell assembly. I wrote elsewhere that this was my assumption
"under typical conditions membranes of all of the excitatory neurons of a minicolumn oscillate in synchrony at a particular frequency. "
"the entire column of excitatory neurons oscillates at a specific frequency."
Essentially the minicolumn & the cortical column (detailed so well by V. MountCastle) is Donald O Hebb's cell assembly.
"In effect, higher amplitude oscillations in layer 2/3 pyramids increases the signal-to-noise ratio of an input at basal dendrites."
It's like saying that the higher magnitude lower frequency tonic oscillations that are common (with greater area effect) provide a high contrast (signal to noise ratio) to the low magnitude high frequency (phasic & high phasic) sensory inputs.
"Sufficiently high amplitudes of oscillations in layer 2/3 pyramids often represent a high level of preparatory attention"
That is why I call the higher magnitude but lower frequency tonic oscillation the ready state, the ground of being, awareness, the edge of criticality. It's a state of being prepared to interpret incoming signal data. It's the formless empty mind in which forms arise.
The locus of consciousness, the canvas of consciousness we can now argue might be based in the Layer 2/3 Pyramid Cells of the minicolumns & cortical columns, and the layer 2/3 pyramidal cells of the hippocampus.
"bursting in axon inputs to the distal segment of apical dendrites of layer 2/3 pyramidal neurons could increase the tonic level of voltage at the soma so that only a small increase in current from a basal dendrite is required to produce a discharge into the axon"
"Up to this point in the present study, the only excitatory inputs to the layer 1 segment of the apical dendrite are axons arising directly from the thalamus within the corticothalamic loop circuit."
"To summarize the main points of this section, the thalamic inputs to the top of the long apical dendrites of layer 5 pyramids can consist of trains of bursts, unlike the inputs to the top of long layer 6 apical dendrites, which almost always consists of the regular firing of single spikes. Each burst is regarded as a surge of current, but its voltage is higher than the voltage of the current produced by a single spike input to the apical dendrite. As these high-voltage surges repeatedly pass down the apical dendrite via the corticothalamic loop, the apical dendrite oscillates at a high voltage. While being fine-tuned by passing repeatedly down the apical dendrite, these oscillating pulses are transmitted to the adjacent pyramids of layer 2/3."
Key Point: high amplitude tonic frequencies create a ready state for high frequency low amplitude sensory input signals
There, they produce an elevated "tonic level" of voltage at the somas, which effectively lowers the amount of voltage at basal dendrite synapses needed to evoke a spike response at the soma.
"This lowering of the response threshold allows inputs at basal dendrites to evoke spike responses in the layer 2/3 neurons when they previously could not. In this way, burst-firing of layer 5 pyramidal neurons can selectively enhance the processing of particular circuits within the horizontal network."
This paragraph below must be read! Their thought experiment was about the image of a dog "appearing" somewhere in between the layer 6, 5, 2/3 cells in that order. They just do not have the concept of computational neural rendering integrated here.
"The circuit effects of burst firing in layer 5 pyramids can be illustrated by the cognitive example of the “hidden image” of a Dalmatian dog in a degraded scene of light and shadows."..."Typically, an observer reports an array of black forms within a white background, but soon the figure of the dog suddenly appears in the foreground."
-
"The present theory claims that the first perception is produced by outputs from layer 6 loops, which produce trains of single spikes to neighboring apical dendrites of layer 2/3 pyramids.
-
"The enhanced spikes of the layer 2/3 neurons are then transmitted to their axons within the horizontal network of cortical processing."
-
"The second perception (the shape of the dog) is produced by inducing a group of layer 5 neurons, representing a part of the whole scene, to fire bursts of spikes."
-
"(...)one can infer that the selection of the dog’s shape and the location of the dog’s shape occurs at virtually the same time"
-
"the supporting circuits for these two events dominate other circuits that might be concurrently active."
I don't agree with the sequence, I would say that the first perception is from the sensory inputs to layer 1 & 2/3 neurons is where the degraded scene of light & shadows transforms into the forms of a dog's shape, location, and the world, and with each oscillating cycle of activity from their upwards and back down again the image & detail is refined. The main factory of rendered perception is going to be in layer 2/3 I think. In the horizontal network of the chattering neurons of layer 2/3. Even so every cell contributes to the rendering with its output. My hunch is that the secondary factory could be the layer 5 horizontal tuft network.
"The tufted top layer of layer 5 pyramidal neuron receives axon fibers from many cortical sources, most of them (90%) as connections from distant locations and less of them (10%) as connections from locations nearby (Hubel, 1982)."
What are the exact mechanisms that cause the opening & excitation or blocking & inhibition of communication synapses when a neuron's phase changes?
"This amplification effect at the soma resulting from facilitating basal dendritic input has been recently described as the Apical Amplification (AA) hypothesis by Phillips (2017)."
"The dendritic tuft apparently represents a separate segment of the apical dendrite that functions as a subthreshold integrator of tuft inputs which has an augmenting effect on synaptic inputs located on basal dendrites."
The Apical Dendrites pass EPSP Excitatory Postsynaptic Potentials via the Thalamic Loop to the Basal Dendrites.
"The idea that synaptic inputs at the apical dendrite tuft can produce a modulating effect on synaptic inputs throughout the layer 5 pyramidal neuron was put forth by Schwindt and Crill (1995)."
The synapses on the back end of a neuron are through intermediate cells modulating (changing) the synaptic efficacy (up or down regulating) synapses in the front end of the neuron that receives sensory input (the basal dendrite).
"They examined the effect of specific channel blocking agents on the current arriving at the soma during iontophoresis of glutamate at a distal site on the apical dendrite and found evidence for amplification of input current at dendritic synapses."
"In addition to increasing the reliability of signals, the spike pattern of a multi-spike burst carries a large amount of information." "based on the n-spike burst code (Elijah et al., 2015)"
"This n-spike coding may be a general property of thalamic neurons
"interspike frequencies could resonate with subthreshold membrane oscillations of a postsynaptic neuron (Izhikevich et al., 2003)"
"which would enable the kind of selective communication between neurons described in the present paper"
"respond to layer 6 axon inputs to set the resonant frequency value of their oscillations, and they also refine that value to less than 1 Hz by repeatedly passing EPSPs along the apical dendrite via the thalamocortical loop."
Are there other ways that synaptic communication blocking might result in neurons being in different frequency bands?
The argument seems to be that the different brainwave frequency bands are like channel separated radio bands, where interference avoidance was accomplished through frequency & spatial separation, somehow the potassium & calcium shuffling that happens when a neuron changes its over all frequency is at the same time exciting synapses to neurons with matching frequency bands & inhibiting synapses to neurons on different frequency bands.
I think Oscillatory physics explains the existence of power bands in brainwaves because oscillations cluster naturally, they bump into each other, change each other, attract each other to a matching frequency through dissipation of signals that results in synchrony, and then they appear to absorb each other, oscillations absorbing oscillations, the greater oscillations have a gravitational interaction with smaller oscillations, so frequency bands form in your brain, but also in space.
What could be the physics based reason for why synaptic communication between neurons that are oscillating at different frequencies would be blocked or inhibited at the synapse?
The splay state effect? Repulsion from an incompatible oscillatory frequency?
I had this idea that perhaps each side of the synapse needs to be in the same oscillating cycle for synchrony to happen. I am guessing this idea doesn't workout because once each side completes its reset cycle it will just sit there until triggered right? I think blocking implies that the receptive channels are not going to open up even if hit with neurotransmitters from a pre-synapse. The cell might be emitting some chemical that causes the channels to freeze up.
a0007z.reverse I proposed the concept of reverse heterosynaptic plasticity with Voltage Gated Potassium Channels.
The other concept is that the ESPS passed from the 6 Layer Apical Tuft to the Thalamus, to the Basal Dendrite are by increasing voltage near the soma through the excitation of some synapses some how shifting the cells charge (in calcium ions) away from the synapses of neurons that are not in sync (like reverse heterosynaptic plasticity) pulling charges away from certain synapses, resulting in the inhibition of synapses that branch to neurons that are out of sync. In a nutshell I'm proposing that the ESPS from the Apical dendrite layer, routed through the thalamus, to the basal dendrite are causing a reverse heterosynaptic plasticity with calcium ions, inhibiting synaptic channels that have grown between neurons that are not currently oscillating at the same powerband frequency.
Besides the novel idea of reverse heterosynaptic plasticity, EPSPs from the Apical Tuft via the Thalamus could potentially also open Potassium channels, inhibiting or hyperpolarizing synapses by leaking charges into the extracellular membrane.
In heterosynaptic plasticity the VGCC Voltage Gated Calcium Channels are opened in synapses that did not fire leading to cascading depolarizations in proximal synapses. However the VGPC Voltage Gated Potassium channels could be triggered by the same heterosynaptic plasticity process, resulting in inhibited or hyperpolarized synapses. Actually this seems so obvious I doubt that I came up with the idea of reverse heterosynaptic plasticity. I just haven't had a chance to look it up yet.
I did not think I would be getting this deep into GABAergic neurons, Chloride Ions, Pyramidal cells, Voltage Gated Potassium channels, Inhibitory Heterosyntic Plasticity & Block Selection of cooperating Cortical Columns before my book was done but here I am.
https://elifesciences.org/articles/37836
"Dendritic Voltage-Gated Ion Channels Regulate the Action Potential Firing Mode of Hippocampal CA1 Pyramidal Neurons"
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2665047/
"EPSPs are also shaped by small conductance calcium-activated K+ (SK) channels. Calcium influx through NMDA-R has been shown to activate SK channels located in the dendritic spines thus attenuating EPSP am- plitude and duration (Adelman, Maylie, & Sah, 2012; Faber, Delaney, & Sah, 2005; Ngo-Anh et al., 2005). Interestingly, SK channels are downregulated following stimulation of synaptic metabotropic gluta- mate receptor subtype 5 (mGluR5) resulting in enhancement of EPSP- spike coupling (Sourdet, Russier, Daoudal, Ankri, & Debanne, 2003). Nevertheless, EPSP-spike coupling in cortical neurons is poorly de- termined by EPSP amplitude and highly dependent on the rate and the waveform of dendritic EPSPs (Larkum, Zhu, & Sakmann, 2001). In cerebellar Purkinje cells, EPSP amplitude has also limited control over cell firing but pharmacological modulation of SK channel and SK-de- pendent plasticity strongly regulate spike firing (Ohtsuki & Hansel, 2018)."
"2.2. Modulation of spike threshold Input-output function may also be altered via modulation of ion channels that control the spike threshold (Fig. 1C). Voltage-gated Na+ (Nav) and K+ (Kv) channels determine the spike threshold (Bean, 2007). Shift of Nav activation towards hyperpolarized values lowers the spike threshold and increases excitability following induction of long- term synaptic potentiation in CA1 pyramidal neurons (Xu, Kang, Jiang, Nedergaard, & Kang, 2005). Down-regulation of Kv1 channels fol- lowing chronic activity-deprivation with pharmacological treatment in hippocampal neurons (Cudmore, Fronzaroli-Molinieres, Giraud, & Debanne, 2010) or following cochlea removal in auditory neurons (Kuba, Yamada, Ishiguro, & Adachi, 2015), lowers the spike threshold and increases intrinsic excitability. It should be noted that in contrast to change in EPSP amplification, spike-threshold modulation is global since it may affect all incoming inputs."
"2.3. Modification of resting membrane potential Change in resting membrane potential (RMP) represents the third manner to modulate input-output function by non-synaptic mechanisms (Fig. 1D). In hippocampal granule cells, high frequency firing induces long-term depolarization (LT-Depol) of their RMP by approximately 8–10 mV (Mellor, Nicoll, & Schmitz, 2002) that is mediated by a protein kinase A-dependent up-regulation of HCN channels. Whereas the up- regulation of HCN channels leads to attenuated EPSP amplitude and therefore to a reduction in intrinsic excitability (see above), the net effect here is however an increased excitability. In fact, the large de- polarization of RMP (8–10 mV) largely dominates the excitability re- duction caused by the attenuation of excitatory synaptic inputs. Here again, this modulation is global as all inputs will be equally affected." https://hal-amu.archives-ouvertes.fr/hal-02363603/document
Off topic note "Nicotinic acetylcholine receptors. Neuronal nicotinic acetylcholine receptors (nAChRs) are pentameric ion channels gated by the neurotransmitter and the alkaloid acetylcholine agonist nicotine. nAChRs are mainly permeable to sodium and potassium, with much less conductance to calcium, and are located on hippocampal pyramidal neurons as well as interneurons (Hogg et al, 2003). Several studies have shown that nAChRs are important for learning and memory in humans and animal models. Blockage of nAChRs in the hippocampus of rats results in significant memory deficits, whereas nAChR agonists including nicotine improve certain types of memory, such as short-term and working memory, in humans (Ji et al, 2001; Levin et al, 2002; Seeger et al, 2004)." "nAChR currents are likely to take part in postsynaptic calcium signaling either directly through their calcium component or indirectly by contributing to postsynaptic depolarization"
On Topic again
"Small-conductance calcium-activated potassium channels (SKs) are widely distributed in the nervous system and are involved in shaping neuronal responses to synaptic stimulation (Bond et al, 2005)
"In hippocampal CA1 neurons, SKs contribute to the after hyperpolarization and modulate neuronal excitability. By allowing potassium efflux after their activation, SKs have the capacity to quench postsynaptic potentials (Faber et al, 2005). In turn, repolarization of the postsynaptic membrane favors NMDA receptor obstruction by Mg2+ ions, which limits further calcium influx. Thus, SKs are part of a negative feedback loop that attenuates synaptic transmission (Ngo-Anh et al, 2005). Indeed, blockage of SKs enhances LTP in the hippocampus and the lateral amygdala, whereas SK overexpression diminishes LTP and impairs learning behaviors such as spatial learning and fear conditioning (Hammond et al, 2006)."
The presynaptic vehicle brakes
"Large-conductance, calcium-activated potassium channels (BKs) also influence synaptic plasticity. These channels are regulated by both calcium and voltage and are localized at presynaptic terminals throughout the nervous system (Hu et al, 2001). Inactivation of BKs increases the probability of neurotransmitter release at synapses between CA3 neurons of the hippocampus (Raffaelli et al, 2004). "... "Thus, BKs might provide negative feedback that moderates signaling through the synapse at the presynaptic side, under conditions of excessive depolarization and accumulation of intracellular calcium." "Specific, non-synaptic voltage-gated potassium (Kv) channels are important for controlling neuron membrane electrical excitability and are localized to axons, somata and dendrites." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1679792/
At a high level NAPOT Revision 4 has caused me to rethink cell assemblies, not just as neurons that are oscillating together, but as neural paths of synaptically active blocks.
I guess I had imagined that neurons from different oscillating cell assemblies might still be communicating synaptically, but now it's more clear that they literally have to reach a shared oscillatory moment, that can happen with bursting, before a message is transmitted between oscillating groups.
Bursting, it might be argued, could be about creating a temporary phase bridge to transfer information from one oscillating group of cells to another oscillating group that is normally out of phase. Something about this is not right about this argument. Let's rethink this part.
"A study by Womelsdorf et al. (2014) recorded cellular activity in anterior cingulate cortex and lateral prefrontal cortex (ACC/PFC) of macaque monkeys during a selective attention task. When the subjects began to attend to the task, neurons in the ACC/PFC increased their firing of brief 200 Hz spike bursts. Burst spikes showed synchrony over long cortical distances; indicating, in particular, that circuits in area 24 (in ACC) and 46 (in PFC)"
"These results support the hypothesis that burst-firing can produce the selection operation of attention by amplifying neuronal activity."
Perhaps it's not as much about blocking or domination as this paper suggests inside a cortical column, from the Layer 6 Pyramidal cell.
Especially since the paper basically outlines how the whole microcolumn oscillates at the same frequency. If that is the situation across the whole neocortex then blocking inside cortical columns seems to have a small role. Also I do not see this paper presenting evidence for how 6th pyramidal cells might effectively block inputs from other neurons that are out of sync phasically. Their paper suggests that the 6th Pyramidal Cells excite the soma's of layer 2/3 cells with EPSP's.
The fact is also that neurons oscillating at 40hz are going to be receiving 200 Hz spike bursts, meaning that this idea of group selection & blocking might apply to group cell assemblies with tonic oscillations. But that is not going to block cell assemblies from receiving out of sync inputs.
This cell assembly selection via blocking might play a greater role synchronizing large cell assemblies that each represent whole cortical columns. It seems that the synchrony inside a cortical column or minicolumn is achieved mainly through EPSP excitatory tuning, and not through domination or blocking, at least with regard to what the 6th layer Pyramidal is doing, as described in this paper.
Instead of describing the Layer 6 Pyramidal tuning as a domination game, perhaps we should describe it as a voluntary enlistment game. Where Layer 6 Pyramids are recruiting cooperative partners with exciting messages.
I think this reconceptualization from "domination" to "voluntary recruitment" makes more sense, because it allows for neurons that are not communicating currently to still be open or listening to new patterns so that they can "hear" the 200 Hz spike burst which could come from anywhere, and would be really hard to ignore, even with inhibited synapses, because of heterosynaptic plasticity + VGCC & VGPC
To be clear I still like the ideas of selection blocking via synaptic inhibition, from EPSPs sent from the 6 layer Apical Dendrite via the Thalamic Neurons cause ions to hyperpolarize synapses. For example if the result of EPSPs directly triggers potassium channels to open or cause other ion activity that hyperpolarizes or inhibits synapses
- If heterosynaptic plasticity triggers synapses that are armed with VGPC Voltage Gated Potassium Channels, then the synapse would become inhibited or hyperpolarized)) but since the neurons in a microcolumn are siloed from other microcolumns & cortical columns, and since the neurons inside a column are in synchrony from Layer 6 Pyramidal Tuning, such a mechanism (referring to blocking) does not seem to be strictly necessary.
- If heterosynaptic plasticity SK Calcium activated Potassium Channels.
- If heterosynaptic plasticity triggers Voltage Gated Chloride CL- channels to open.
Instead voluntary recruitment of oscillatory synchrony via EPSP signaling seems to work here, and it also works fine with the description of the physics of how oscillations outlined in the book Sync by mathematician Steven Strogatz.
What I am saying is that blocking probably does happen this way, on a small scale, temporarily, as a weak & passive short term effect. I have doubts about whether it could be the main driver of selection between cooperating cell assemblies or cooperating microcolumns & cortical columns.
With columns having electrical isolation + dendritic membranes being naturally leaky + entropy there is enough working against the physics of synchrony in the first place to make blocking via synaptic inhibition seem like it plays a small role in selecting groups of cell assemblies to work together.
I do think (inhibition or blocking) is a real effect, but just not the driving effect.
I think selection (or oscillatory grouping), at least as described in this paper, is mostly from excitatory synaptic potentials EPSPs, not from this interesting metaphor of neuronal blocking for dominance.
While writing this I remember David Eagleman saying something like "Imagine the brain is a cooperation of rivals" or "The human brain runs on conflict" - Eagleman, maybe not Mr. Eagleman. Maybe not. Just humor.
Conjecture: Whether you describe the activity as voluntary cooperation patterns, or as a conflict between choices, this may just be a matter of characterization, a made up story either way. Cooperation with one idea is a conflict with another idea.
Of course domination through excitation is a viewpoint one can have.
I feel like the real motivation of the 6 layer pyramidal cells is a lot more about energy dissipation & oscillatory synchrony, and not about selection via blocking.
a0007z.interneurons a0309z.interneurons
Alternatively: Block selection of whole mini columns & cortical columns, via Thalamic Matrix Neurons triggering GABAergic Inhibitory Interneurons makes a lot more sense.
Blocking for cell assembly of whole columns between cortical columns via Thalamic Matrix Neurons communicating with GABAergic Inhibitory Interneurons makes more sense, but the focus shifts from united individual neurons with individual frequencies, to a focus on selecting (uniting & dividing) whole cortical columns to work together. Since a neo cortex column tends to oscillate at one frequency the focus on blocking individual neurons with signals from 6th pyramidal EPSPs may not be the most important factor in cooperation between cells.
GABA GABA
"GABA is a general inhibitory neurotransmitter, which is sensed at GABAergic synapses by ionotropic (GABA-A) and metabotropic (GABA-B) receptors. GABA-A ion channels consist of five subunits forming a central pore that is permeable to negatively charged chloride ions (Cl−; Baumann et al, 2001). Cl− influx through GABA-A receptors hyperpolarizes mature postsynaptic neurons expressing appropriate Cl− transporters and inhibits synaptic transmission."
This could be why cocaine addicts (Putin) think their bad ideas (like Russia invading Ukraine) are good ideas. Less GABA CI- influx, less inhibition. (humor)
"Repeated exposure of rats to cocaine reduces the amplitude of GABA-A-mediated synaptic currents and increases the probability of spike initiation in dopaminergic neurons of the ventral tegmental area"
"GABAergic and pyramidal neurons of deep cortical layers directly receive and differently integrate callosal input"
My statement: Perhaps most of the blocking & inhibition & selective cell assemblies are accomplished with GABAergic inhibitory interneurons and CI-influx (Chloride Ions) If that is the kind of blocking that the people writing the Neural electric tuning paper were thinking about then another big picture is emerging.
- selection of cooperating neural columns might be a bigger deal then block based selection between neurons inside a single column.
- We would need to look for a possible link between the 6th layer pyramid cell & its EPSP's that are sent to basal dendrites of OTHER cortical columns via the Thalamic Matrix Neurons.
- We would need to find a possible link between the 6th layer pyramid cells & the GABA activated Chloride receptors.
- Or show a different path that simultaneously stimulates 6th layer Pyramidal & GABA interneurons, and shows that they can function in concert, together. Allowing GABA Interneurons to fulfill the role of blocking whole columns of the Neocortex that are oscillating on a different frequency power band, and grouping columns of the cortex that are on the same resonating frequency.
"parvalbumin (PV)-(...) Stimulation of callosal fibers elicited monosynaptic excitatory postsynaptic currents in both layer VI pyramidal neurons and γ-aminobutyric acidergic (GABAergic) interneurons immunopositive for the vesicular GABA transporter and PV"
Okay so it looks like Pyramidal Neurons and GABAergic interneurons are simultaneously stimulated (presumably by Thalamic Matrix Neurons)
"The data show that pyramidal neurons and GABAergic interneurons of deep cortical layers receive interhemispheric information directly and have properties supporting their distinct roles."
FIGURE 7 A1 "Different temporal integration of synaptic responses evoked by callosal stimulation in pyramidal neurons and interneurons. (A1) Representative current clamp recordings (single sweeps) showing EPSPs evoked by 4 stimuli at different frequencies in a pyramidal cell and a GABAergic interneuron"
"Temporal Integration of Callosally Evoked Responses Is Different between Pyramidal Cells and GABAergic Interneurons All the data presented so far suggest that layer VI pyramidal cells and interneurons are likely to integrate the callosal input within a different temporal time window. "
"4 callosal stimuli delivered at 40–200 Hz produced temporal summation of evoked EPSPs. When pyramidal cells were recorded (n = 9), the probability to elicit an action potential riding on EPSPs was inversely correlated with increasing ISIs (Fig. 7A1). Importantly, even at gamma frequencies, the stimulation could trigger action potentials. When interneurons were recorded (n = 8), a similar correlation was observed but shifted toward significantly lower ISI"
"These results demonstrate that pyramidal neurons integrate the callosal information over a longer time window than interneurons."
"(...)challenge this view. We show here that PV-positive GABAergic interneurons of deep cortical layers of RSA/frontal cortex receive synapses on dendrites and somata and are directly activated by the contralateral hemisphere."
"We also found that pyramidal neurons of layer VI are monosynaptically excited by the callosal fibers. This finding also fills a gap in literature (...)"
"Therefore, the information conveyed by the callosal fibers would then spread from layer VI neurons toward the thalamus synchronizing a bilateral cortical-thalamic loop, as well as toward several other cortical sites."
https://pubmed.ncbi.nlm.nih.gov/16829551/
Slightly off topic or off thread but this is perfect for NAPOT 1 with 5 types of discharge are defined below.
"we explore in this paper the ability of the PRC neurons to amplify the output voltage to current input at selected frequencies, known as membrane resonance"
"In this way, we identified 5 different types of discharge (Figures 1B–E,G–K).
(1) Late-spiking regular neurons (RS): At just-suprathreshold, these neurons show a slow ramp depolarization before the onset of their spike trains, with a consequent delay of the first spike. At more sustained depolarizations, they are characterized by a persistent tonic or slightly adapting firing. In line with this, their ISI distribution is linear and almost parallel to the X-axis (red dots in Figure 1L). Also, their ISI-CV2 relationship shows a cloud of dots very concentrated and close to each other at a low CV2 (about 0.1) (red dots in Figure 1M).
(2) Stuttering fast-spiking neurons (FS): At just-suprathreshold, these neurons fire trains of high-frequency spikes (30–50 Hz) separated by variable periods of silence. At more sustained depolarizations, they are characterized by a persistent high-frequency (50–100 Hz) tonic firing. Likewise in RS neurons, their ISI distribution is linear and almost parallel to the X-axis and their ISI-CV2 relationship shows a cloud of dots very concentrated and close to each other at a low CV2 (about 0.1). However, the FS dots (orange) can be distinguished from the RS dots (red) because they are shifted to lower ISI values (Figures 1L,M).
(3) Adapting neurons (ADP): Adapting neurons typically begin their spike trains at a short latency following onset of a depolarizing current step and accommodate strongly. Due to adaptation, their ISI distribution is linear, but with a higher angular coefficient than RS and FS neurons (blue dots in Figure 1L). Also, their ISI-CV2 relationship shows a quite dispersed cloud of dots (blue dots in Figure 1M) with a higher mean CV2 (about 0.4).
(4) Bursting neurons (BST): Bursting neurons are characterized by spikes that occur in a stereotyped pattern consisting of a cluster of 2–3 action potentials riding on a slow depolarizing wave and followed by a strong slow afterhyperpolarization. After the burst, their firing generally becomes regular. Therefore, their ISI distribution is not linear but starts with shorter ISIs (green dots in Figure 1L) and their ISI-CV2 relationship consists in a rather compact cloud of dots (corresponding to the regular firing) accompanied by two or three more dispersed dots (corresponding to the burst) (green dots in Figure 1M).
(5) Irregular neurons (IR): Irregular neurons show a random and unpredictable firing pattern. Their ISI distribution is dispersed and not linear (black dots in Figure 1L) and also their ISI-CV2 relationship consists in a dispersed cloud of dots (black dots in Figure 1M). They have a mean CV2 similar to that of adapting neurons (about 0.4), but their nonlinear ISI distribution uniquely characterizes them.
On topic: Resonance between Pyramidal cells & GABA Interneurons
"A regular 50sec-long ZAP current input with linearly increasing frequency from 0 to 15 Hz was applied to test for resonant behavior of pyramidal neurons (Figures 2A,E) and GABAergic interneurons (Figures 2A′,E′) at a membrane potential of −70 mV. Resonance appears as a peak in the voltage response at a specific frequency (Fres) (Figures 2A,A′), that is absent in non-resonant cells (Figures 2E,E′). As a consequence, resonant cells show a peak in the impedance-to-frequency relationship at Fres (corresponding to the dashed vertical line in Figures 2B,B′), whereas a clear peak is not detectable in non-resonant cells"
"Accordingly, the phase shift-to-frequency relationship and the complex representation of the impedance differentiates between resonant (Figures 2C,D,C′,D′) and non-resonant (Figures 2G,H,G′,H′) neurons, through clustering of positive values in resonant neurons. The percentage of the resonant pyramidal neurons and GABAergic interneurons measured in the superficial and deep layers of areas 35 (A35) and 36 (A36) of the PRC is shown in Table 3. Overall, the majority of perirhinal pyramidal neurons (77%) and GABAergic interneurons (54%) were resonant and were equally distributed throughout the PRC, without a clear prevalence in a specific area or layer (Table 3), suggesting that resonance could be very important for the oscillatory synchronization and integration of the neuronal activity in this region. Also, the resonance strength (Q–70) and the frequency of resonance (Fres) were similar in pyramidal neurons of A35 vs. A36 and of superficial vs. deep layers (Figures 3A–D). Comparable results were obtained also in GABAergic interneurons (Figures 3A–D). However, we found that the Q–70 was significantly different between pyramidal neurons and GABAergic interneurons regardless of their location (Figures 3A,C)."
To make a long story short we don't need to find a method to explain how ESPS from 6th layer pyramid cells can, via thalamic loops can block communication between neurons that are oscillating in different power band frequencies because the GABA Interneurons are receiving the same signals as the VI layer Pyramidal Cells and are in resonate synchrony with them which means they are able to act as a single sensor transmitter system that blocks, gates or inhibits communication between cells, or mini columns or cortical columns allowing for a selective assembly of cells for given tasks that you might wish to perform.
https://www.frontiersin.org/articles/10.3389/fncel.2021.703407/full
Why it makes sense that channels stay open to neurons of a different frequency
Slow wave potentials and inhibitory waves resulting from action potentials may also have the effect of re-coding information to be compatible with other areas of the cortex that are oscillating at (lower frequencies in the higher layers) different frequencies.
If you want to transfer sensory information from a part of your brain that is moving very fast to track very fast things, to a part of your brain that is thinking much more slowly & deeply about issues, then the high frequency sensory information has to be converted down into a message that the slower frequency part of you (in the upper layers) can process. So you have different frequency bands, for different sensory data tracking, and you have these different kinds of action potentials to up regulate and down regulate the frequency of information patterns, so that different areas of the brain that are operating at different frequencies can participate in processing (considering) and reacting (choosing an action).
One odd conclusion that came to me while reading this paper on the Tuning of Oscillations is that the only real goal of the brain is oscillatory equilibrium. One funny implication to consider is that if life ever became too easy humanity might evolve backwards into trees. Why would we move if we didn't need to move, and if we don't need to move why do we need brains? So if we use sentient self aware networks to make life really easy for mankind, we could devolve into trees, I mean, at the rate of evolution, so that could be on a time scale of millions to billions of years.
We evolved because the circumstances demanded it.
"Hippocampal synaptic inhibition is mediated by distinct groups of inhibitory cells. Some contact pyramidal cells perisomatically, while others terminate exclusively on their dendrites." "In contrast, dendritic inhibition may control the efficacy of afferent inputs both by suppressing the generation of dendritic calcium spikes and by limiting depolarization due to excitatory synaptic events (Wigstrom and Gustaffson 1983; Kullman et al., 1992)." https://www.sciencedirect.com/science/article/pii/S0896627300801014#:~:text=In%20contrast%2C%20dendritic%20inhibition%20may,et%20al.%2C%201992).
"Distinct synaptic properties of perisomatic inhibitory cell types and their different modulation by cholinergic receptor activation in the CA3 region of the mouse hippocampus"
"Perisomatic inhibition originates from three types of GABAergic interneurons in cortical structures, including parvalbumin-containing fast-spiking basket cells (FSBCs) and axo-axonic cells (AACs), as well as cholecystokinin-expressing regular-spiking basket cells (RSBCs)." "The perisomatic region is defined as the domain of the plasma membrane which includes the proximal dendrites, the cell body and the axon initial segment (AIS; Freund & Buzsaki, 1996). This region is targeted by three types of inhibitory neurons in cortical areas, namely by the parvalbumin (PV)-expressing fast-spiking basket cells (FSBCs) and axo-axonic cells (AACs) as well as by the cholecystokinin (CCK)-containing regular-spiking basket cells (RSBCs). The basket cells innervate the somata and proximal dendrites (Blackstad & Flood, 1963) whereas the AACs target the AISs of pyramidal neurons (Somogyi, 1977)." "Basket cells are multipolar GABAergic interneurons that function to make inhibitory synapses and control the overall potentials of target cells. (this quote is from wikipedia on basket cell)" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2916217/#:~:text=The%20perisomatic%20region%20is%20defined,Freund%20%26%20Buzsaki%2C%201996).
GABA regulates resonance and spike rate encoding via a universal mechanism that underlies the modulation of action potential generation
"(GABA) causes regularly firing hippocampal CA3 neurons to bistably switch between spiking and quiescence (inhibition state), converts graded discharge-to current relationships to have abrupt onsets, and induces resonance. Modeling reveals that A-currents enable these GABA-induced transitions. (...)this transition sequence (universally) underlies the modulation of AP (Action Potential) dynamics."..."In simulated networks, synaptically controlled AP dynamics, permits dynamic gating of signals and targeted synchronization of neuronal sub-ensembles. These results advance the systematic understanding of AP modulation" https://www.biorxiv.org/content/10.1101/206581v1.full.pdf