Musings on 5th Generation Warfare, #6

Continued from here.

A step increase in human intelligence was my goal, and I spent nearly 2 weeks evaluating possible approaches. From the beginning, it was clear that the most generally useful of savant capabilities were the rarest, e.g. the kind of people who absorb huge amounts of information, keep it all in working memory, and can dictate entire textbooks without error. That would be a goal, probably one of many as we learned about the varieties of possible intelligences, but likely lesser versions, all varieties of pattern matching and calculating abilities due to a more direct connection with raw sensory feeds, or levels just above that, and with better memory.  Patterns drive all thinking and discovery, so I was sure improved pattern detection, deeper pattern matching and analysis would be generally useful.

Investigating the phenomena of savants, they seem to be the result of dis-inhibiting regions of a normal brain, with extreme cases also having overgrowth of brain cortex, especially right hemisphere, due to damage in other regions. So how to arrange benign versions of that? Electrical and magnetic stimulation was beginning to reliably produce changes in mood, artistic and mathematical ability.

One of the thoughts I had had while looking at genetics was enhancing the amount of hybrid vigor.  That had happened naturally in polyploid plants, which had duplicated their sets of chromosomes, all of which could afterwards evolve independently.  All of our major cereals are polyploid because they are therefore naturally more vigorous and more easily develop resistances to rusts, blights and bugs, but now also because it is easier to breed them with multiple disease resistances.

However, making a polyploid human probably wouldn’t work.  First, polyploid fish, amphibian and reptiles exist, but examples of mammals are rare or non-existent, mostly liver and heart cells.  Second, fused cell lines normally lose or rearrange chromosomes, so producing a polyploid animal that way doesn’t work. Third, we already had an example of the female X chromosome as a warning that adjustment is necessary, also from the plant species.  As females have 2 Xs, one chromosome must be disabled in every cell to prevent over-production of the products specified on the X chromosome.

However, that mosaicism of females was also a good clue, as the chromosome that is disabled is randomly chosen and is not active in any of the cells that descend from that cell through embryological development, so entire organs may be one parent’s or the other’s cells.  Nevertheless, one of the reasons women are thought to be more robust then men is the hybrid vigor available from 2 different alleles of genes on the X chromosome.  One of which is disabled in every cell, meaning that 2 genetically different sets of cells inhabiting one body can compensate for each other’s metabolic deficiencies!  Perhaps even across organs. It seemed to me that a better mix of parent cells would produce better compensation, at least.

Genetic mosaics are not so rare, formed by fusing two gametes in utero or a placenta shared between fraternal twins or by the mother’s cells crossing the placental barrier and continuing in her child, so a lot was known about them.  Beyond avoiding the obvious, don’t mix sexes, otherwise tetragametic, 2 eggs and 2 sperm, fraternal twins fused, people have been discovered and were perfectly normal otherwise.  That woman had had children with all of her genetics, so the cell lines were thoroughly mixed. This had been replicated in mouse and other animals, although most of those had been genetically close.  Those animals were normal, so the process didn’t seem to harm the resulting individual. Judging from the female X example, hybrid vigor in general, I thought that genetic closeness was the reason the animals hadn’t been more better.

My hypothesis was connected with Kirschner and Gerhard’s thinking in several ways. First, genetics don’t specify the body or organ, they are a recipe for building bodies and organs.  It all happens in the embryo and as the body grows.  Growth and normal operation of the physiology is all controlled by the timing of biological processes stepping the genetic state machine in each growing and dividing cell through the 1000s of combinations of genes enabled and disabled in the order needed to build an animal. The initial brain, the last of the organs to be completed in youth and one that continues to produce new cells and connections so long as a person lives, is apparently very sensitive to changes of many kinds, as people who seem normal become insane at a certain rate, and the much increased rate of new autistics said somethings in our environment were a problem.

Timing is everything in the biological state machine, it is the reason for critical periods : some genes need particular kinds of stimulation from the environment to continue the sequence needed for a normal brain to develop.

A second point of K&G was that the body has favored producing organs with more capacity than needed in normal conditions, if that capacity can be had for relatively low cost. Therefore, evolution selects for the ability to adapt to new conditions, e.g. produce more muscle and supply the required nerves and blood vessels.   They term those ‘exploratory processes’, ones that have a local physiological goal, e.g. forming a capillary from artery into a tissue with too-low oxygen tension and on to attache to a venule, or making the capillary form an arteriole to supply yet more blood, or a nerve cell’s attachment to muscle or other nerve cell and later deciding if it should continue that attachment based on feedback from the cell it attached to.  That is all driven by internal states of each cell, coordinated through messages with adjacent cells and the physical medium they are suspended in.

The brains’s various critical periods suggested that those adaptation processes went on throughout at least adolescence in mammals.  Different timings in one brain should have interesting effects, as each cell line worked through their own critical periods and attempted to produce their genetic’s version of excess capacity and pruned their networks by slightly different criteria in the target cells.

The major argument of K&G is that evolution has produced a biological system and genome which has evolved to make further evolution easier, what they call ‘facilitated variation’.  The deeply conserved base structures, the ‘core processes’, have been selected for the types of linkages and ordering of expression to decrease the probability of fatal genetic errors and thus make genetic change more probably useful.  The last few chapters of the book deal with the embryological compartments established by the gastrula? stage and which guide differentiation.  These lay out the mechanisms in development that have been constrained so as to allow the expression of new phenotypes to be reliably produced by genetic change, evolution favoring further evolvability because it would most rapidly populate a new niche environment with its own descendent species, all inheriting a common set of traits that make further evolution faster, less costly in lost lines of genetic innovations.

So my hypothesis is that a normal brain develops according to the same schedules in all of the neurons, and that allows normal inhibitions of brain region by brain region as well as development of normal synapses.  We know from the research on savants that some of that inhibition reduces adult capacity in return for being able to generalize.  Reasoning in generalities is not always accurate, to say the least. I thought that having multiple genetics in one brain with somewhat different timing would reduce the amount of inhibition. That could increase the number of cells and perhaps even produce new structures.  In any case, these were cheap experiments, and quick if we used mice.

So through my various contacts and funds, I set in motion the preliminary research to test the idea of mosaics having greater-than expected intelligence.  We quickly repeated the mouse-mouse and rat-rat mosaics between wild strains of each with different laboratory strains. Those showed definite improvements, but when more than 2 sets of different parents were used, the animals were both larger and smarter.  For instance, combining an asian mouse with a European mouse with a laboratory strains produced a mouse 50% larger than a normal mouse and smarter than the smartest rat.

The early lab work was using shotgun tactics, trying everything because we had no clue what would work.  Those wild-assed ‘what if’ experiments continued in almost every laboratory, a classic exploratory breadth-first search algorithm, all of us keeping each other up to date on the latest results.  We spent half our time in the forums exchanging thoughts, trying to bring all information to bear on each question posed, each result reported.  The intensity of this work, the interest from all segments of society was phenomenal.  Many good suggestions came from left field, as usual, from people of no particular background, just general interest.  This was humanity improving itself.

For standard combination work, we quickly settled on the major technologies. Using pluripotent stem cells derived from clone lines allowed rapid repeated experiments.  Once we developed the technology that allowed complete mixing of all cell lines in all organs, progress was rapid on all fronts, mostly because the results were so uniform, all mice or rats were a smooth mix, no cell lines dominating some organs.  We automated the testing of animal intelligence using the traditional bar press apparatus, but forcing the animal to detect the pattern of how many presses on one followed by how many on the other were necessary to receive the pellet.  After they got to 9/10 correct sets of presses on one pattern, we switched them to another.  Mice never got to the point where they understood to start experimenting as soon as the old pattern wasn’t working on any except for a simple reversal of which bars produced pellets, although they did learn more complex patterns and extinguished the previous pattern faster than normal.  For the same mix of 2 wild and one lab rat, that ‘pattern reversal set’ learning was achieved by rats for more patterns of 5 different bars pressed 5 different counts, a first for the species and as good as small birds.  You should not need more hints about how little we understood of intelligence.

The major difficulty was the size of the Tessels, as we termed any mosaic with 4 or more parents.  After we saw that the larger brains were most differently organized, and the animals much more intelligent, in fact their intelligence was often startling on particular problems, quirky overall, we pushed for more iPSCs in every embryo. More parents produced better intelligences than more iPSCs per parents, tho we didn’t see other effects on bodies. 8-parent embryos are still our current limit, and those didn’t often have uniform mixes of cells in the early days.  The animal doesn’t end up 8 times larger, but often twice as large, with twice the size of the head and brain.  That was too large for a normal animal’s birth canal, Caesarean section was required, and a foster mother able to deliver enough milk.  So more manual work, more technicians, more planning.  Less cheap, but compared to the average high-energy physics project, the cost of our step function in human intelligence was peanuts.

That work was moved into our new company, Tessellated Genetics, funded by friends of mine.  We immediately opened our lab notebooks to the network, confident we were going to make $ with this venture, and wanting the technology to develop as rapidly as possible. We did make $, these were relatively cheap experiments, and after the initial work most of it was funded by producing and selling the standard combination mice and rats and the clonal lines of cells used to produce the induced Pluripotent Stem Cells that everyone used in behavioral and neurophysiological studies of the phenomena and in developing their own variations.

That was production-line work, once the base technology of injecting the iPSCs into blastocysts was worked out. Harvesting fertilized eggs, raising them to blastocyst in cell culture, maintaining clonal lines in cell culture and producing iPSCs was normal operations, barring disease or genetic aberration in the mouse or cell culture lines.  Even injecting blastocysts was automated, tho not the implanting in the female rats and mice, nor the handling of animals, care and cleaning and setting up tests, running sets of animals through them, recording data.  It took a lot of people to run this operation, distributing the work was crucial to rapid success.

During this time, we also pioneered doing research for individuals.  That is, this was all open-source, the lab notebooks were open for all to see throughout.  We got many comments and suggestions from interested people, some of whom could fund the research they suggested. We always had new students in our labs, we ran a lot of courses, and those students needed projects.  As soon as they had mastered the basic techniques, and when we had the ability to properly supervise them, we funded them by individuals sponsoring their work and getting the social credit.  Details of the experiment were often decided between funder and student, “James H. Hamilton’s experiment by <researcher> reported in Cell”, etc. references were common in all of the laboratories after that, and this became a common way of funding things after my reformation had diminished the old order. Open Science was the new norm.

It must be admitted that those studies produced a lot of negative results.  However, this had a very positive effect on science, as we published all of them in peer-reviewed open journals.  They were often the first such published failure, prevented others from wasting their time, and taught our young scientists to do experiments as well as positive results would have.  Even better, we often ended up doing more research to explain the failures. Those, along with the positive results and unexpected results more than made up for the cost of the failures.  Knowledge is hard, this is the most efficient way to produce it reliably : add more randomness to the ideas being tested.  The restricted range of ideas that could be considered was the major failure of science in the government-funded approach.

Economic analysis in future years showed that this approach also prevented the second major failure of government funding, the fact that it did research before its time.  That is, one can build a particle acclerator costing $10B or a Tokamak costing $5B per year, or wait for accelerators-on-a-chip and Cold Fusion.  In the interim, the money put into those large projects would have produced many times the return if distributed randomly over research proposals.  Generally, the research turns out not to be the important element of a proposal, the researcher’s experience is the important variable in the progress of science. Science need more breadth-first searchers.

Among other exciting results, we saw that higher initial intelligences hard more upside in the quirky intelligences produced. Dog Tessels were much smarter than the smartest wolf ancestor, and put cat Tessels in the shade.  The monkey work was shortly in the future, and confirmed the trends.

At this point, assuming I had the answer for intelligence, I next had to consider the organization of a society with a minority of Tessels. Best case, all Tessels would be very intelligent and knowledgeable in more areas and would see deeper patterns, more probably for some time we would produce people with lesser capabilities, and perhaps some would be autistic, there seemed to be more ways to make autistics than super-intelligent individuals, just looking at the statistics for normal populations.  However intelligent and uniformly intelligent, omniscience would not be an attribute, so they were important elements for constructing the civilization I envisioned, but could not be the whole answer.  If they were used incorrectly, or even if the usual unexpected side-effects of a new capability were too unexpected, they would be the new disaster.  So many proposals of this type had been.

It seemed to me that a society of Tessels + humans had too many instabilities.  There wouldn’t be many of them relative to normals, all of their offspring would be normals. It would be important to keep them dispersed, otherwise they could enable more centralization if put into research, planning or logistics positions. Also, humanity has not done a very good job of making and keeping allies and using all of their capabilities for improved civilization, so to the extent that Tessels were different, they would be treated badly by some humans, and that could easily become another social and cultural divide, of which humanity already had too many.

Even best case, I had to assume some would be very different. Alone, they were not enough to correct the current system, however intelligent. But a few of Frankenstein bent, and we could have a permanently estranged minority as an unfortunate complexity.  Even if we stopped producing them, the problem would not likely disappear quickly, and if they were so valuable we couldn’t stop, maybe so valued we wouldn’t stop if the ratio was one super-genius per 1000 annoying, grating, nearly too strange to allow to live on their own, and tending to perform some bad but not horrific crime or social outrage,  it could become a big problem.

Additionally, I thought we humans didn’t have a good handle on how to transfer information, knowledge and wisdom mind to mind.  I had been encouraged that the Massively Open Online Courses were competing in this area and finally were developing the measures to know what worked, whether there were different styles of learning and how a teacher had to adapt. However, I hadn’t heard anything much from that yet (but should have, so my info stream was inadequate), for example that certificates from one were worth more in the job market than those of another, so no improvement on traditional universities was apparent. I didn’t see much guidance for the problem of educating Tessels.  And it was a problem : how to assess a quirky intelligence? How to train it?  What tasks?  What datasets? These new varieties of humans would be major investments, optimizing the return was important.

Another thread was that I had not yet solved the problem of men’s sexual drive producing centralization.  This would take some thought, as Tessels and the required social organization for that technology was another complexity to be fitted into this base structure for the next civilization.  More reading, more thought, a lot of ‘what iffing’.

Along with this, the overall K&G story was one of an evolutionary system that had evolved itself for faster evolution.  Our society had done the reverse in every area it could control and centralize.  Again, I saw that preventing that was key to an optimal progress in civilization. Channeling men’s sex drives was a key element, preventing that and the insecurity it produced throughout society from powering a need for power.

Working through the interlocking conundrums took weeks.

In psyops, the message is the op.

*Generalissimo Grand Strategy, Intelligence Analysis and Psyops, First Volunteer Panzer Psyops Corp.  Cleverly Gently Martial In Spirit

Continued.

Advertisements

3 thoughts on “Musings on 5th Generation Warfare, #6

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s