current-stuff

Recurrent neuronal nets with sheet inhibition make stable, coherently sub-networks. They are the on top of the neurons, and can be see as the "neuronal letter" of the "neuronal alphabet" (Yuste).

My idea is that there is a new kind of biology, the structure and function of the neuronal ensembles. They are a kind of animal.

Levin tentatively suggests "Ideas are alive". We merely have a substance chauvinism.

Braitenberg, Donald Hebb, Rafael Yuste, György Buzsáki, others: The Neuronal Ensembles.

More complete doc

---

Hebb: Cell Assemblies, The Organisation of Behaviour 1949

• theoretical psychology
• Influenced Dan Dennett

focus: causality structure

If A goes to B, then B follows A. The cell assemblies together with the proposed local learning rule, Hebbian Learning is called Hebbian Theory. Neurons that fire together, wire together - the slogan is somewhat misleading. It is the neurons that cause each other to fire that wire, it takes 2 timesteps to do that.

Thought Pump:

• Activate the halo of the assembly by increasing the excitability.
• By decreasing the excitability, the assembly would narrow down into one of its centers.
• This depends on a threshold device, something that regulates the thresholds (excitability) of all neurons in question "area"?, column?, whole cortex?, 'sheet wise', claustrum?, basal ganglia? thalamus?, not figured out.
• It should select the best-connected sub-center (given ideal inhibition, see below).
• This move then, could be done at the speed of 2 neuron timesteps, and it could be a parallel search process.
• Jumping from thought to thought, creating a 'train of thought', hence 'thought pump'.

• If the assemblies are 'settled' and "don't move" anymore, perhaps this could be interpreted as the 'best reachable interpretation' for the system.
• This remains to be purely conceptual afaik. But is a cool idea to have in any case.

Since you are scrolling through already, check out this work on visualizing much of the activity in a zebrafish larva:

---

Assemblies

---

Moshes Abeles: Synfire Chain

And so forth.

Perhaps the details don't matter.

---

Reverberating across brain areas for multisensory integration

[Pulvermueller: Semantic grounding of concepts and meaning in brain-constrained neural networks] [A neuroanatomically grounded Hebbian-learning model of attention–language interactions in the human brain]

---

• The dragons of old turned out to be dinosaurs
• So the spirits of old turn out to be the software that animates the world. (Josha Bach, Valentino Braitenberg).

---

What is brain software?

An operating system of a biological computer that makes a magical interface to it's user (also person).

The Magical Interface

• Is the benchmark for a good interface. As in: "This tool is like an extension of my mind".

Perhaps It allows an ad-hoc crafting and debugging of pieces of technology until something that sticks is found that serves some purpose until a replacement is found.

Sevant-syndrome indicates that the capacity of this technology is seemingly infinite, the bottle-neck being how much time and effort the person is engaging with the activity in question.

• The developing brain has 4 years + some months to make a 4 year old
• The brain performs a software synthesis task, this must incorporate substantial design

---

The synaptic overproduction and subsequent pruning seems at first glance a general fit for evolutionary variation and selection.

Synaptic pruning:

• at least 50% of synapses disappear
• neurons that act jointly appear to maintain synapses between them
• 'the brian is sculped by activity'

critical periods:

WIESEL TN, HUBEL DH. SINGLE-CELL RESPONSES IN STRIATE CORTEX OF KITTENS DEPRIVED OF VISION IN ONE EYE. J Neurophysiol. 1963

---

• genotypes, phenotypes?
• meme engine wizards

link

(Colors merely aesthetic).

Gene: A piece of information that replicates and is instantiated in the genome as "genome snippet" (the second meaning of the word gene).

Knowledge: A replicator which causes itself to be stable in an enviroment, also called it's niche. I.e given a niche it is a replicator.

Adaptation: The effects that a knowledge has, given a niche, that causes it to be a replicator.

Adaptation is allowed to become intricate, with many sub-parts contributing to a system of effects. I would say the means of survival and replication - called the extended phenotype (Dawkins 2016).

This typically involves an organism, instantiated via development (Ontogenesis) in biology, but it is not limited to the borders of an animals skin.

Indeed it is not limited to biological substrate at all, as in technology like beaver dams and even more insubstantially, in animal behaviour. And propensities for learning (Baldwin), psychological developmental plans (A. Sloman), etc.

The phenotype is extended, actually this is another thing sayable with constructor theory. All possible effects that genes can have are allowed to be phenotypes.

Adaptationism: The bio philosophical position that holds that design in biological systems is created by natural selection.

Natural selection: a approximate construction, whose substrates are populations of replicators and whose (highly approximate) constructor is the environment. Marletto Constructor Theory of Life.

Is a mechanism that is sufficient for creating design, i.e. adaptations.

And is the only known mechanism for creating design in biology.

Natural selection needs a population of variants of replicators, from which the ones with highest fitness (inclusive genetic fitness in biology) are selected. I.e. the ones with best design, the best effects that caused them to replicate.

For this reason, any adaptation can only be an adaptation (i.e. display design), if it caused by a replicator for which the variants are possible. 👈

---

An attribute is a set of states of system.

Adaptation, the sense of design has to be caused by a replicator. Information with causal power (C. Marletto, S. Walker).

In constructor theory of information - the permutations have to be possible.

Something like that:

For all attributes of an adaptation a, there exist permutations of the information variable Σ, that causes the variants of a.

Not sure what the variants are, has to do with the kind of universality the system achieves?

The interpreter of the genomic code (Ribosomes in the case of genes) have to have 'a certain universality'.

---

If a natural selection theory of brain software design is true, there should be a 'genome-like' information medium.

1. The information variable Σ, is a set of attribues where the task of peforming the permutation over Σ and the replication task over Σ are possible.
2. The attributes of an information variable are called information attributes.
3. An information medium is a substrate whose attributes constitute an information variable.

(C. Marletto 2014, ct-life):

??

1. Do the information attributes of the genome-like neuronal meme medium correspond to the neuronal code?

Everything the neuronal interpreter can do, the neuronal memes need to be able to say.

---

Here I'm drawing what I think Dawkins has in mind:

It should also be noted that the message system used in the nervous system, […] is of an essentially statistical character. In other words, what matters are not the precise positions of definite markers and digits, but the statistical characteristics of their occurrence, i.e. frequencies of periodic or nearly periodic pulse-trains, etc. Thus the nervous system appears to be using a radically different system of notation from the ones we are familiar with in ordinary arithmetic and mathematics: instead of the precise systems of markers where the position - and presence or absence - of every marker counts decisively in determining the meaning of the message, we have here a system of notations in which the meaning is conveyed by the statistical properties of the message. […] This leads to a lower level of arithmetical precision but to a higher level of logical reliability: a deterioration in arithmetics has been traded for an improvement in logic.

John von Neuman - The computer and the Brain

Pentti Kanerva - Sparse Distributed Memory

hyperdimensional computing:

• H, the hyperspace is vast.
• 'The blessing of dimensionality'
• All hypervectors are roughly same distance apart.
• There is a tiny region in H where inputs similar to the address lie.
• Encountering inputs that are similar to the address is astronomically to be random.
• 'programming with embeddings', can model semantic spaces, 'concept particles'
• and structure via VSA operations: Datastructures, trees, sequences, FSA, etc.
• programming in superposition (in the classical domain)

The address decoder neuron:

Hearing the voice of a person, say 'Bob', always sounds like the same voice, even though the particulars of the sensor data must be quite variable.

In Hofstadter's terms, we make an analogy between the instances of hearing Bob's voice. We understand/categorize/make the analogy / analyze the situation / find the essence of the 'hearing Bob's voice situation'.

Whatever represents such an internal situation is quite robust to noise.

The Reader Centric Cell Assembly

1. They share a downstream reader.
2. They fire together in a time window given by the integration time ("tau") of the reader.

Neural syntax: cell assemblies, synapsembles, and readers György Buzsáki

Attention Approximates Sparse Distributed Memory

---

György Buzsáki

The Brain From the Inside Out 2019

---

Valentino Braitenberg

https://en.wikipedia.org/wiki/Braitenberg_vehicle

---

P. Kanerva, “Hyperdimensional computing: An introduction to computing in distributed representation with high-dimensional random vectors,” Cognitive Computation, vol. 1, no. 2, pp. 139–159, 2009.

Sparse Distributed Memory, Pentti Kanerva 1988

---

Peter Pesic

https://mitpress.mit.edu/9780262543897/polyphonic-minds/