Two important things came together this week. I am seeking feedback on my TENTATIVE plan for how to connect them.
Last Sunday, in the dialogue on the seven challenges to humanity I have been talking about, the greatest excitement came when I inserted a SECOND slide (see the two slides attached).
When I worked at NSF, I led the creation of a radically new research topic, "COPN,"
That led me to develop a new integration of our understanding of how the brain works, in a FUNCTIONAL way, as a source of higher intelligence and consciousness,
unlike narrower work which would usually focus on just one small PART of the brain aimed at specific narrow tasks. I have summarized that new view, informally, at:
In the links there, you can see photos of Walter Freeman, Karl Pribram and Karlk Lashley -- the three greatest SYSTEMS neuroscientists ever. (SYSTEMS neuroscientists are those who study how the brain works as a WHOLE system, with a higher universal learning capability, what we ask for now in Artificial General Intelligence AGI.) COPN looked for 50-50 BALANCED projects, advancing our understanding of how brains work BUt ALSO connecting that to breakthroughs in AGI. The deep learning revolution which is sweeping the whole world right now, originated in one of the grants which came from COPN, as shown at
https://www.werbos.com/Mind.htm. What we already learned in the first round of COPN already points clearly to new extensions of AGI far more powerful than what most policy "experts" imagine possible today.
I am deeply delighted that Frontiers in Neuroscience has announced that they will publish a kind of SEQUEL to the Freeman issue. More precisely, they sent me more information today
including:
========================================================================================================
Fees and waivers
You will find additional information on article types and publishing fees linked. We consider waiver requests on a case-by-case basis (with a typical response time of up to a week). Institutional agreements may also partially cover costs, mainly in Europe. I’m also available to answer any questions your contributors might have on fees.
Deadlines
The next deadline is the summary deadline, which is currently set for November 17th 2024. The manuscript deadline is set for March 7th 2025, but Confirmed Contributors can set personal deadlines that we can revise together.
==========================================================================================================================
I was also delighted when David Wack of Buffalo agreed to serve as editor-in-chief of this special issue. Myself, I believe that I am too old and too overwhelmed by the responsibilities you see on the FIRST slide attached.
However, I have agreed to serve as one of the co-editors serving under David, and I hope that a few of the others I bcc here might consider joining us, and joining in some follow-ups.
I am glad they were willing to consider my proposed scope and announcement for the special issue (v2 DRAFT attached), and plan to send out their final announcement relatively soon.
IF the special issue (and other follow-ons) get far enough, I really hope that NSF EFRI (or UNESCO?) will consider issuing a follow-on to COPN, using this as evidence that there IS a community interested and serious enough to get us all further.
================================================
BEYOND AGE, another reason why I should be co-editor, not editor in chief, is that the community needs to know very clearly that this issue will be open to all serious coherent viewpoints on the unavoidable, important issue of whether the human mind is JUST a pattern of the brain, but ALSO a kind of symbiotic combination of brain and something else, which I call "soul" (with apologies to those who use that word with interpretations very different from the definition I use in my own work).
When I was a program director at NSF, I tried hard to hide many of my personal views, because the world of research needs networks of inputs from MANY viewpoints, and I needed to avoid creating a false impression about some kind of official party line. (I remember when colleagues could not even guess whom I would vote for.) But as co-editor, I feel I have a right and responsibility to INCLUDE, as ONE of the viewpoints, some of what I think I have learned since COPN. That would mean a position paper by me, saying a bit about the history, but mainly developing FOUR KEY THEMES from my personal viewpoint:
(1) An HISTORICAL/FOUNDATIONAL PART
This would mainly reflect the kind of background described above, which is already worthy of a vast new mainstream research effort. (In fact, as I type this, I could even see a reason to include THIS or the published version as part of our next round of provisional patent filing!)
For example, I might mention my very first meeting with Karl Pribram in Radford, Virginia, circa 1990, arranged by Sam Leven, whom I bcc. Karl began by saying "you neural network modelers are basically crazy and out of touch with reality. I have read all that stuff from Steve Grossberg and his followers in computational neuroscience, developing Hebbian models based on Steve's doctrine of The Neuron Model, using differential equations where outputs traveling along axons are all that really matter. He needs to learn about FIELD effects, and about flows in all directions. Until your people start producing models LIKE THAT, I will not take any of you seriously." Oh, did I smile at that!!! That was the beginning of a long collaboration Certainly Werbos and Davis cited lots of work by Buzsaki, studying how electromagnetic field effects are crucial even for "spike sorting", for processing of real brain data, even to figure out what neurons actually DID output at various times (as our paper analyzed, proving effects well beyond what that old neuron doctrine considered possible).
YET Pribram's final scientific book, Brain and Perception, still relied heavily on electromagnetic field effects to explain what was most definite in all of his data (unless you count the radiant smiles he often received from beautiful women when HE smiled at them... part of MY stunned first person experience).
His conscience and his sense of wonder caused him to ask for a kind of "quantum appendix" in that book, by Yasue and Jibu, whom I later had VERY extensive contact with.
ARE QUANTUM effects (still within the scope of quantum electrodynamics, QED, the best-known sector of modern quantum field theories) important in a functional way even to good old mundane brain dynamics? Can we find out at least that much in new hard core science?
That question was very visible in the COPN solicitation, and I learned a lot from our debates, including a lot which cries out for follow-on work. More and more, it did seem clear that quantum computing effects or entanglement ACROSS neurons (via QED physics, not accounting for ideas like global qi entangling us with life beyond the brain) was a fringe idea, advocated by believers full of hope but less and less hope of realistic possibilities. But quantum computing effects WITHIN the neuron seemed plausible then, and still today, deserving more mainstream research. Long before Stuart Hameroff built up a huge devout following for HIS version of quantum physics and even gravity in the brain,
I funded Michael Conrad of Wayne State University, who passed very tough NSF review in one of my three core program areas (Adaptive and Intelligence Systems), and who developed models of quantum associative memory (QAM) of interest and new potential even now (albeit without the gravity, so far as I can tell). If the full version of this gets published in Frontiers, I will add some material which belongs in our next provisional patent filing. The unexplained high energy efficiency of biological neurons compared with anything known to mainstream electronics today led to a massive research program at DARPA, which Robert Kozma and Hava Siegelmann participated in. The DARPA program never replicated that high energy efficiency, but QAM and a related quantum design concept may be important both to understanding how biology does this and how to replicate the same.
II. SITUATIONAL UNDERSTANDING FOR HUMAN SURVIVAL AND HUMAN POTENTIAL
Here is where I have a duty (an overwhelming dharma reinforced by the strongest meditation) to present a new, heretical viewpoint.
Humans (and human groups even more) have a tendency to go to extremes, go to the crazy left or the crazy right, in part because of group loyalty effects and related economic pressure.
And so, perhaps >90% of all humans today believe EITHER that "soul is a total hallucination" (as explained in Hebb's otherwise very great book The Organization of Behavior, one of the two foundations of the entire neural network research field -- especially the Grossberg school) OR THAT when we die we just teleport to some other place in some other world or some other body.
Myself, I am one of the VERY few people on earth who really understands the most important foundations of modern quant
um field theory, as used in fields ranging from depth psychology to working device design (but in philosophy only by philosophers who meet my definition of first-order sanity given in
https://link.springer.com/article/10.1007/s41470-019-00038-z ).
Thus I AM an extremist on the issue of believing in physical reality. The links above explain how I now resolve the huge apparent contradiction between believing in Hard Core Einsteinian Realism (HCER) and in believing in some kind of SOUL. Most efforts to bridge this gap simply do not work, in hard real logic... but with time, I have come to accept an alternative worldview, in which the human "soul" is basically what we are connected to in a larger intelligent system, which I call our "solar system noosphere." I do not really call that entity "God," because, like us, it is finite; I feel resonance with people like Teilhard de Chardin who call it "Gaia" or "Omega", but I prefer the nickname "Amaterasu" (which I first learned from Mari Jibu!), because it is not all-wise, is not an adult of its universe-spanning species, and is not limited to only the earth.
THE SECOND SLIDE in the first attachment comes from Werbos and Davis, but depicts how ANY intelligent system at the mammal level or higher actually works. (It depicts what follows from mathematics, following the approach of my draft paper Approximation4, which I hope Fronteirs might allow as a supplementary materials file.)
OUR souls are, roughly, like CELLS in a larger brain, a larger intelligent system. SOME cells do not get the feedback (aka backpropagation, cathexis, qi) needed to justify the continued use of resources in that brain; just as "garbage collection" subroutines in AI recycle resources which have not delivered enough value to the brain, SOME human souls end up like what is discussed in the Book of Esdras (one of the Apocrypha in the Bible), or like the souls dissolving into powder in the great Disney animation "Coco".
(In truth, I once visited one of the two leading Buddhist monasteries in Korea, on Chicken-Dragon Mountain.
Earlier I saw a debate between the leading Zen and Tibetan teachers in US.
The Zen advocated "no mind" or nonexistence, while the Tibetan aimed at "mindfulness," an OPPOSITE goal deeply engraved in his full mind.) I saw an old female monk who semed to be achieving great success in HER path, a Zen path... almost dissolving into powder before my eyes. THAT was one hell of a first person experience! (it reminded me of creatures dissolving in huge pain with a huge smile in the important novel Voyage to Arcturus.)
BUT MANY of our souls end up like memory interneurons (following Grossberg mathematics, ONE of the cell types we tracked in Werbos and Davis).
I view them as the kind of souls depicted in work by Annie Besant, in books I saw in the shelves of Mahatma Gandhi's apartment in Mumbai. They look like immobile statues to us, UNTIL we touch/activate them with OUR energy. MEMORY cells.. activated only when signals from OTHER types of cells activate them.
In truth, one famous Eastern teacher taught that people actually DIVIDE UP after mundane death. Perhaps SOME parts of your soul get recycled, SOME (or the totality?) get transformed into memory cells, while OTHERS transform into giant pyramid cells of the noosphere, like the left hand side of the slide. Bernie Baars -- a founder of the OTHER (not Hameroff) annual scientific conference on consciousness -- in HIS chapter in Freeman and Kozma (Freeman's final summing up of how brains work), argued that the "Global Workspace of Consciousness" is actually a layer of cells in neocortex, as shown in this slide. I interpret that to mean our normal "conscious" awareness of reality in our mundane brain is exactly the outputs of these giant pyramid cells, considered as a single set of numbers, the vector R-hat. THAT picture fits into part I, new research in the mundane brain, not requiring these controversial possibilities for soul.
But for SOME of us... WE may leave traces of ourselves still alive in the noosphere, as blocks of the giant pyramid cells of the noosphere.
THOSE OF US whose souls receive enough feedback from the noosphere -- which come to us as sensations of qi capable of perturbing brain and soul both -- will persist after mundane death, IF the feedback was strong enough. (I am very glad and grateful to have received such feedback.) That feedback is governed by the mathematics of modulated backpropagation, driven by the connections we make in the FORWARDS direction from OUR soul to OTHER CELLS AND TO the noosphere as a whole.
And so... I think of the second slide as a kind of model of our noosphere as a whole, and of our place in it. Those of us who never connect simply get evaporated at the end.
The rest of us go through a very painful "alchymical divorce", either in one quick shock or in a step-by-step process (as depicted in the great novel
https://www.amazon.com/Passage-Novel-Connie-Willis-ebook/dp/B0030P1WSQ/ ). Our future life depends on the quality of connections we made. AS IN A BRAIN, connections to the whole are essential, both to us AND TO THE ABILITY OF OUR ENTIRE NOOSPHERE LIFE TO SURVIVE THE COMING CENTURY!!! And DIVERSITY of cells is essential, as well as their ability to work together through diversity. (If all cells had the same input-output relations, it would a stupid and ineffective brain indeed!) More and more, this includes the essential growth of new connections through and to the internet, and CONNECTING networks aimed at different subgoals, as we see in brain design (e.g. basal ganglia, an important part of the model,
discussed even in important work by Grossberg, albeit in need of additional modulation).
Later, this work led to the granting of a patent this year (October 15) (the top link at https://patents.justia.com/inventor/paul-j-werbos) which shows how these kinds of extensions are of crucial importance in increasing the chances of human species survival, as in the first slide attached. That patent assumes only firmly proven QED physics (
https://drpauljohn.blogspot.com/2024/02/concrete-proven-but-little-known.html). It is already enough to tell us how to "see the sky" much better than we can today, with implications for missile defense, cybersecurity and other very urgent applications.
But it could also be applied to new sensor technologies, to build networks capable of detecting nuclear signals from beyond the earth, which may be far more important than we
III. Fundamental levels of intelligence in Mammals and Birds
Long ago, Skinner and his school postulated that there is only one fundamental KIND of brain, one universal learning rule described in mathematics describing the highest learning capabilities (and "consciousness") operating across ALL kinds of animals.
Many years later, great seminal papers by M.E. Bitterman demonstrated several basic qualitative TYPES or LEVELS of intelligence. His most classic scientific paper back in the 1960s in Scientific American worked out experiments demonstrating several TYPES of intelligence, like mammal, bird, reptile, amphibian, fish. For many years, the serious neural network intelligence work
like COPN aimed at understanding and replicating precisely the
MAMMAL level of general intelligence.
Later, however, after Mirror neurons were discovered, people like Freeman and myself developed a theory of how the evolution of mirror neurons resulted in the evolution of human language, the very foundation of how humans became "top dog" (more or less) on earth. (E.O. Wilson often suggests that ants and bees are the real top dog, a topic deserving careful dissection.)
And so... in my papers in Neural Networks in the 1990s, I illustrated levels of intelligence WITHIN the mammal class... (1) the early level, without mirror neurons:
(2) early primates capable of reconstructing the sensory input and motor output of OTHER creatures they see, in THEIR vast associative memory, used in their ability to "learn from memory';
(3) Humans, a kind of half-evolved species (as described in famous quotations form Kopnrtad Lorenz https://www.brainyquote.com/quotes/konrad_lorenz_392014), who have developed an ability to act out or DANCE their own personal pat memory and to SEE another human dancing, to extend the power of their mirror neurons still further; (4) "sapient" creatures, like what humans are evolving TOWARDS, which would be born with something like what I call "first order "sanity" or the honest use of words and other formal symbols. (Humans are not BORN
being honest, either in words or in other symbols, either to others or to themselves).
NOW HOWEVER, I have probed new research both on some mammals and on birds, which strongly suggests we need research to figure out the GREATER diversity of fundamental levels than what we knew before.
While relaxing, I read a novel "The Idiot Gods" (now called "The Orca's Song") by Zindell which talks about new research into the intelligence of cetaceans.
The names and titles he cited can be tracked through scholar.google.com. They point to very serious new results, strongly suggesting that many cetaceans have evolved very fundamental and important learning capabilities beyond what any (or most?) humans possess! Through the parahippocampus and paralimbic system, they seem to extend the mirror neuron system in a way which allows them to "see through many eyes at once", and fuse different viewpoints. This reminds me of those higher Upanishads (see Hume's collection) which discuss the transition from seeing through the two eyes of the self to seeing through the eyes of the Self. In NSF EFRI reviews after my retirement, I saw signs of new empirical neuroscience work beginning to track down the neural circuits which may be responsible, across different fundamental ways of using mirror neurons. Just as the basal ganglia in rats learn from a memory structure which might be called verb-object or verb-object-modifier, the human level and cetacean levels may make more use of a SUBJECT component in those verb-object record, in a way which is fundamental to understanding such higher mammal intelligences.
But beyond that... even birds may show great diversity, maybe even more than mammals.
exhibiting the kind of language behavior which Freeman and I had studied in humans. That suggests that they too must possess some kind of mirror neuron system,
though perhaps using different wiring, like what humans possess. It was especially amazing to see the sociology of the relations between a penguin city and a human city of about the same kind, with similar hinterlands and trading networks. But new research on songbirds suggest that they too might -- or might not -- possess a kind of high-level intelligence which Bitterman and Wilson would both appreciate. Do those small songbirds possess SOME kind of collective or "hive" intelligence like what E.O. Wilson talks about, which Bitterman WAS able to demonstrate in insects in his later papers in Science? What about raptors, from magpies to ravens to crows to eagles or even owls? Certainly the native Americans of Alaska and Canada have many stories of intelligent and cooperative behavior in such species, which should be better understood.
And, if all such creatures have "souls" (as I would expect from the theory of Part II), we should not just laugh at all the stories of indigenous peoples everywhere about contacts
with such creatures -- even though SOME stories must of course be based on confabulation, a phenomenon very well known in humans, consistent with the theory depicted in
my second "challenges" slide.
============================================================
=====================================================\
ADDED LATER: IMPLICATIONS FOR HUMAN EXPERIENCE
x
This morning, at the time of the golden dawn, I found myself again of thinking about the great Disney animation,
The Lion King, which a young boy rightly named Leo brought my whole family to watch a few weeks ago.
The movie depicts a "circle of life" which is important and powerful for US. I THOUGHT I already saw that movie many years ago, but
I am grateful for how it revived and brought new life to my own personal understanding of how our lives are really working now, especially as we get older.
But my wife reminded me this morning: there were no GRANDPARENTS in that movie. Where are WE aging people in the circle of life?
IT DEPENDS ON THE SPECIES IN A VERY FUNDAMENTAL AND IMPORTANT WAY.
Fundamental...
of how intelligence and mind really work in the brains of mammals and birds. I hope they will accept this (perhaps with constructive modifications) in the new special issue of Frontiers in Neuroscience
which I am part of. (I previously sent you links to the announcements from the journal).
It actually is fundamental and important to ask "Where are the grandparents in this circle of life?"
It depends a lot on the species. A few years ago, Walter Freeman and I watched VanDer Posts videos on the life cycle of the Kalahari Bushmen, and arrived at a stark view of the LEVELS of intelligence in mammals: (1) basic, like what rat and mouse have; (2) basic mirror neurons, like what monkeys have, raising their intelligence in a very fundamental way, letting them store
an encoded memory of what ANOTHER primate was doing, so that they could learn from a larger memory database including records of what OTHERS experienced; (3) human, where we can ACT OUT our past experience in dance, so that another human can watch us and encode/assimilate our PAST experience. Humans are not born with language like what Chomsky and Pinkert theorized.
Rather they learn it as a kind of dance of words, made logical only after culture TEACHES us to use unnatural modern languages like English try to force logic into language.
which GO ONE (or more?) level further, adding a new way of USING mirror neurons, related to the idea in
more ancient Upanishads of the greater Self seeing through many sets of eyes at once.
There is also evidence of fundamental levels of design in birds as well. I learned a lot about this through direct observation this year,
towards the South Poleand the far north, from Alaska to Acadia.
AGING AND THE LEVEL OF INTELLIGENCE/MIND is a crucial factor in deciding how the circle of life actually works for us mammals.
of you want to know (though the most advanced new knowledge could buy us many decades of healthier life). Very briefly, we mammals all possess a kind of telomere mechanism,
which kills aging cells unless they are revived, but which also prevents cancer. TO DO BOTH requires energy -- more food. EVOLUTION addresses the tradeoff: when is it "worth it"
(as judged by natural selection) to keep an old creature alive longer, when it requires a cost in terms of more food?
THERE IS ONE OBVIOUS INSTANT REPLY (ONE PART of what is obviously a multivariable story):
Evolution will favor keeping grandparents alive IF THEY contribute enough to the survival of the family or tribe.
When I first started learning about cetacean intelligence and human aging, I naturally went to the web and saw tabulations of how long different species live.
There is a HUGE variation, even among creature otherwise similar in body type. And there is a relation (rough, modulated but very strong correlation) between
qualitative level of intelligence and lifetime. Humans live a lot longer than dogs and cats, even though all are social group hunters. Humans have an ability to
pass on important useful information and experience to the next generation.
In our recent visit to Acadia (guided by people working at the Maritime University of Maine) we learned how SOME key species have longer lifetimes than dogs and cats
let alone rodents: puffins, ravens, eagles, harbor dolphins.... all about 30 years. All RELATIVELY intelligent and sociable. But the "para" cetaceans live much longer.
Orcas are rightly quite famous (as depicted in Zindell's great novel the Orca's Song), but there are more peaceful dolphins we see far less (they are smart enough to stay away from humans in most places, except for Toba dolphins like what we met in Japan a few years back).
IF NIH wanted to do research to extend HUMAN longevity, instead of doing just research on mice as a model, maybe they would visit these dolphins! There actually is a major joint international Pacific research program, strongest in Japan (Toba!) and US, which might help. Furthermore, as in the old
NSF COPN research program , this would help enrich our understanding of the kinds of neural network structure which could also be used to upgrade the power and empathy of artificial general intelligence, even beyond the mammal level implementation of emerging quantum AGI for which my
new patent was recently approved. (Unfortunately, the term "QAGI" sounded so exciting to many software marketers that they often use it for much older and weaker ideas. I go by the original definition in the patent, and the papers incorporated in it.)
