a cell's perceptual universe

[Bruce Nevin 2018-09-11_14:32:53 ET]

(This has been delayed since 09-07.)

Eetu Pikkarainen 2018-09-07_11:10:47 UTC –

Bruce Nevin 2018-08-31_08:08:00 ET –

Digression: It is the cells that constitute the loop that do the reorganizing by making and breaking connections. For some reason, failure to control a signal which necessarily cannot itself be a variable that is perceived and controlled by the cell has an effect on inputs that are controlled by the cell, so that it starts randomly changing its synaptic connections. Maybe we could figure out ‘what’s in it for the cell’ by modeling rates of firing under conditions of failure to control. Conflict is an obvious case, and in conflict the loop increases output to its maximum capacity. Perhaps sustained high rate of firing is the condition that the cell tries to control by randomly reorganizing.

(EP: Is that a quote from some publication?)

BN: No. This question has long been on my mind: “What’s in it for the cell?”

It refers to the magic of embryology, maturation, learning, and healing generally in any multicelled organism, but here particularly to the cells that constitute the neural embodiment of a hierarchical perceptual control system.

The most general answer that occurs to me is that there is survival benefit in constituting part of a stable environment, where the environment is predominately made up of cells. (It may seem too general to say, with Thaddeus Golas, “We are equal beings and the universe is our relations with each other.”)

Consider one nerve cell. It has extended parts of itself to specific locations relative to other cells. In a motor neuron that location might be a meter away. It accumulates and releases electrically charged atoms (ions) along the length of those extensions and through elaborately structured membranes brought into close proximity (the synaptic gap) to such membranes on the like extensions of other such cells. The frequency of this accumulation and release (the rate of firing) is essential for the functioning of the control system which the cell helps to constitute, and therefore it (the frequency) cannot be a variable that the cell itself can ‘perceive’, since any perception once recognized might come under control, and if the cell controls the rate of firing conflict will inevitably arise sometime between the cell and the organism of which it is a part.

From this naturally come speculations about side effects of our control which are imperceptible to us. Bill quite justly inveighed against any notion of a “superorganic” system of which humans are constituents. I would want to see the sensors, input functions, output functions, and effectors, he said. On the analogy of the relation of the nerve cell to the neural signal, we must be incapable of perceiving such matters, though we might be able to deduce them–perhaps to our peril. Now there’s an interesting science fiction scenario.

We return you now to your regular programming.

(Last time I did this, Bill said he felt like the floor and ceiling had been pulled away and then put back.)

[From Erling Jorgensen]

Bruce Nevin 2018-09-11_14:32:53 ET

BN: This question has long been on my mind: “What’s in it for the cell?” …[with accompanying digressions and speculations]…

Hi Bruce,

EJ: I am looking forward to exploring such ideas with you at the IPCT Conference. You are raising the right kinds of questions. I have become interested in the structures and properties of Comparators in the nervous system, as potential sources for the meta-monitoring that control loops need. But reorganization happens for perceptual input functions too, not just output functions. So there either needs to be a deeper structure for reorganization, or mechanisms for sustained error to have broader effects.

EJ: By the way, I will be staying at the same hotel as you are, (in Morton Grove, I think), so hopefully we can arrange some good exchanges of ideas. See you in about a month.

Erling

Confidentiality: * This message is intended only for the addressee, and may contain information that is privileged and confidential under HIPAA, 42CFR Part 2, and/or other applicable State and Federal laws. If you are not the addressee, or the employer or agent responsible for delivering the message to the addressee, any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this in error, please notify the sender immediately and delete the material from your computer. Thank you for your cooperation.*

Please also note: * Under 42 CFR part 2 you are prohibited from making any further disclosure of information that identifies an individual as having or having had a substance use disorder unless it is expressly permitted by the written consent of the individual whose information is being disclosed or as otherwise permitted by 42 CFR Part 2.*

“Bruce Nevin” (bnhpct@gmail.com via csgnet Mailing List) csgnet@lists.illinois.edu 9/11/2018 2:34 PM >>>

[Bruce Nevin 2018-09-11_14:32:53 ET]

(This has been delayed since 09-07.)

Eetu Pikkarainen 2018-09-07_11:10:47 UTC –

Bruce Nevin 2018-08-31_08:08:00 ET –

Digression: It is the cells that constitute the loop that do the reorganizing by making and breaking connections. For some reason, failure to control a signal which necessarily cannot itself be a variable that is perceived and controlled by the cell has an effect on inputs that are controlled by the cell, so that it starts randomly changing its synaptic connections. Maybe we could figure out ‘what’s in it for the cell’ by modeling rates of firing under conditions of failure to control. Conflict is an obvious case, and in conflict the loop increases output to its maximum capacity. Perhaps sustained high rate of firing is the condition that the cell tries to control by randomly reorganizing.

(EP: Is that a quote from some publication?)

BN: No. This question has long been on my mind: “What’s in it for the cell?”

It refers to the magic of embryology, maturation, learning, and healing generally in any multicelled organism, but here particularly to the cells that constitute the neural embodiment of a hierarchical perceptual control system.

The most general answer that occurs to me is that there is survival benefit in constituting part of a stable environment, where the environment is predominately made up of cells. (It may seem too general to say, with Thaddeus Golas, “We are equal beings and the universe is our relations with each other.”)

Consider one nerve cell. It has extended parts of itself to specific locations relative to other cells. In a motor neuron that location might be a meter away. It accumulates and releases electrically charged atoms (ions) along the length of those extensions and through elaborately structured membranes brought into close proximity (the synaptic gap) to such membranes on the like extensions of other such cells. The frequency of this accumulation and release (the rate of firing) is essential for the functioning of the control system which the cell helps to constitute, and therefore it (the frequency) cannot be a variable that the cell itself can ‘perceive’, since any perception once recognized might come under control, and if the cell controls the rate of firing conflict will inevitably arise sometime between the cell and the organism of which it is a part.

From this naturally come speculations about side effects of our control which are imperceptible to us. Bill quite justly inveighed against any notion of a “superorganic” system of which humans are constituents. I would want to see the sensors, input functions, output functions, and effectors, he said. On the analogy of the relation of the nerve cell to the neural signal, we must be incapable of perceiving such matters, though we might be able to deduce them–perhaps to our peril. Now there’s an interesting science fiction scenario.

We return you now to your regular programming.

(Last time I did this, Bill said he felt like the floor and ceiling had been pulled away and then put back.)

[Martin Taylor 2018.09.12.13.43]

              [Bruce Nevin

2018-09-11_14:32:53 ET]


BN: No. This question has long been
on my mind: “What’s in it for the cell?”

                  It refers to the magic of

embryology, maturation, learning, and healing
generally in any multicelled organism, but here
particularly to the cells that constitute the
neural embodiment of a hierarchical perceptual
control system.

                  The most general answer that occurs

to me is that there is survival benefit in
constituting part of a stable environment, where
the environment is predominately made up of cells.
(It may seem too general to say, with Thaddeus
Golas
, “We are equal beings and the universe
is our relations with each other.”)

Yes. That is, in effect, the answer that I came up with at a very

different scale. (see attachment). The cell is likely to survive
longer if the organism in which it lives is stable, which means that
it controls well.

SideEffectLoops.18.08.27.pdf (1.38 MB)

···
What follows is a long dissertation analyzing a conceptual approach

to this issue, using an approach different from that in the
attachment, but related to it.

The role of the cell in the organism's ability to control means

nothing to the cell’s actual operation. It’s an evolutionary thing.
When Powers initially considered reorganization as a random
re-weighting and reconnection process, he realized that to be
effective within an entire individual would take times comparable to
the age of the Universe, not of the age of the individual. It was
only when he discovered the “e-coli” process that he was able to
take the random aspect of reorganization seriously.

Similar arguments apply here in spades, but the solution might be

different. The first organisms presumably controlled very few
variables, and used chemical rather than neural signalling. The
archetypal “e-coli” for PCT controls just one, and its model
oversimplified biological bacterium perceived only which end-to-end
direction of its body points up a gradient of some scalar field such
as concentration of a nutrient in the liquid in which it swims. Its
descendants might perceive and control a second variable, such as,
say, the side-to-side or up-down variation of the same
concentration, which would allow the mutated bacterium to choose the
optimum direction in which to turn rather than making a tumble in a
random direction.

From there, multiple systems might be developed using a variety of

sensors and output mechanisms, but so long as the reporting was al
done chemically, each different signal type would need to use a
different chemical. The situation would be analogous to the early
stages of TV, when each transmitting station had to use a frequency
band that differed from its neighbours, but could use the same band
as stations a long way away. Initially there were 13 bands (channel
1 was soon discarded), and as the demand for variety increased, UHF
bands numbered up to (I believe) 85 were added. But the electronic
properties of the atmosphere varied by the day and by the hour, and
sometimes in Toronto a station from Houston or (on one occasion) San
Diego might override the signal from Buffalo, and sometimes the
signal from Buffalo would be so weak as to fail to produce a
recognizable picture in the “snow”.

In the much-mutated complex single-celled descendants of the

hypothetical original super-simple “e-coli”, on occasion chemical A
from the upper north-west corner of the cell might leak into the
southern sector, where that chemical was used for an entirely
different purpose (a side-effect disturbance to the southern control
systems). Or it might get depleted and fail to provide the signal
variation required to enable effective control of its variable.

In the case of TV, this kind of problem was resolved (among other

problems such as energy wastage) by not using a broadcast medium,
but instead providing point-to-point services such as cable or
directed narrow beams. In the biological case, these might initially
have been membrane-encased tubes that kept the chemical spatially
separated. Such tubes would have varying concentrations of the
signalling chemicals along their lengths, as the signals progressed,
and they would have to converge on hubs where the signal values
represented by those concentrations were produced or used.

Varying chemical concentrations often are accompanied by electrical

potential variations, as also happens when one material moves with
respect to another. Since electrical effects can be transmitted from
point to point much faster than can material concentrations, it does
not take much imagination to see the value of using the tubes as
electrical rather than material transmission conduits, without
changing the interchange of chemicals, using the electrical signals
to stimulate the emission of the old-style chemicals from a reserve
supply held at the far end of the tube, and for the receiving hub to
limit its sensitivity to those chemicals to small areas around the
tube-ends. I have just described a synapse.

At this point, we are talking about a "cell" that contains

internally a lot of spatially distinct structures, all of the same
general character, all contributing in one way or another to the
maintenance of the "cell"s control of various internal properties
that depend on its local environment. These internal structures,
which we might now call “archeo-neurons”, swim in a chemical
environment that has some spatial variation of chemical
concentrations – a “hormonal soup”. What we have been calling a
“cell” now might be considered to be a multi-cellular organisms, or
approaching being one.

In this proto-organism, what's in it for any one of the

archeo-neurons? Only that proto-organisms in which they promote good
control are more likely to survive to produce descendants than are
those that detract from the overall ability of those that don’t.
There’s nothing in it for any particular archeo-neuron, from this
point of view, but the later existence of connection structures in
which they participate cooperatively depends largely on whether they
do or do not participate in effective control structures.

In the attachment, I talk about social systems, and about

specialization of roles in sufficiently large communities of
organisms. In the above, I concentrated on one kind of
specialization, but for the development of the archeo-neurons there
must be other specializations, such as structures that produce each
particular kind of chemical used in signalling. They are analogous
to the farmers, cart-wheel-makers, storekeepers, metal-smiths, and
so on, who prove different services that allow the other kinds of
social structures to survive.

Whichever kind of structure we choose to examine carefully, the

others are there to sustain it. A specialized “farmer” would not
exist if everyone grew their own food, for example. It is because
there is a farrier to shoe the farmer’s horse that he can plow as
big an area as he does, and because he has a customer for his food
that he needs the farrier. Specializations exist and survive because
of their position in networks of use and be used. The same is true
of the kinds of cell in an organism. There may be nothing in it for
any one cell, other than that it survives longer in a stable
organism whether it knows it or not. But there sure is a lot “in it”
for that organism’s kind, and its likelihood of existing , and
existing in the same or similar kind of network, down through the
generations.

Martin