Rough Doublings of Rate of Progress in Natural History

[jordan179]

I was thinking about the longevity of life on Earth, and I saw this interesting pattern. All the dates are very approximate, and I'm aware that my rounding may be creating a pattern where none really exists. Anyway, roughly

Comments

[zaimoni]

At least in simulation (genetic algorithms on neural networks and similar), teachable intelligence is an adaptation to an environment so chaotic that instinct is lethal. Once the the environment stabilizes, the "common facts" get swapped from intelligence to instinct in a few hundreds of thousands of emulated generations.

Of course, the simulated genome in question is somewhat smaller than a minimal real-world virus (either biological or computer). But the qualitative results are consistent with what is inferrable about African climate and geography since the first Meditterraean sea dehydration (Messinian Salinity Crisis, estimated time 5.6-5.8 million years ago).

[firstashore]

You've got it roughly right, but I'll put in a few comments:

It's no secret that life has been evolving in complexity in a roughly exponential way. That's why you can pick these arbitrarily "significant" events and they sort of work out - even though your dates are arbitrary they represent an underlying stochastic pattern.

There is a video I posted on my LJ a few months ago that illustrates this very well. It's actually an argument against Intelligent Design but it works for your purpose here too. It's a series of simulations of life and how natural selection works on a series of clocks (the ID watchmaker argument). The simulation behaves in exactly the same way as you've described here. The first improvements are minor and take a very long time to evolve, but the later stages accelerate exponentially in complexity. See it here.

One problem with the way you're doing this, though, is that you're only halving it each time and thus missing some extremely important events. Particularly in the Devonian (~400Ma), which is probably the

[zaimoni]

The difference between reptilian and avian/mammalian brain plans is not merely a matter of efficiency. The avian and mammalian brain plans share form-parsing visual structures that the reptilian brain plan doesn't have a trace of. The disabling of these visual structures while retaining the reptilian ones in the human brain, results in blindsight: no qualia of vision whatsoever, but full ability to react to motion, color, and light intensity changes in the objective visual field.

[jordan179]

The surprising thing, comparing mammals to avians, is that the avian brain plan appears to be superior, not only to the reptilian (so is the mammalian) but also to the mammalian. In other words, birds are more intelligent per gram of brain per gram of body than are mammals. Corvids and pscatticines have roughly the brain-body ratios of cats or dogs, but they are roughly as intelligent as apes or ceteceans.

The reason seems to be that avians have bare-bones, ultra-efficient organ systems in general, under the evolutionary paring knife of the need to reduce mass to be able to fly. The cost avians seem to pay for this afdvantage is that they have a (relatively high) random chance of dying in any given year from major organ failure: in other words, they have less protective redundancy built into their organs, compared to mammals

[zaimoni]

When the tradeoff is efficiency vs functional redundancy, there is no clear "superiority" just from the plan; a larger context is required.

Object demonstration: the potentially Third-World electric grid in the U.S. The electrical generation and transmission reserves are cost-inefficient, which didn't matter too much with vertically integrated regulated monopolies. The cost-inefficiency does matter under reregulation -- leading directly both to the 2001 California power crisis, and ensuring that the Sun again can take out the electric grid in the Northeast U.S. And a nation-wide operating procedure of normally running high-voltage power transmission lines at 105%-110% of design capacity (in careful rotation of course, so that they don't melt).

• I understand that the Aug 14 2003 blackout is contestable in this regard, but there was a K9 magnetic storm two days before, and significant magnetic storm activity at the exact moment when 850MA of current decided to go West instead of East. I don't think the failsafes failing at Detroit

[zaimoni]

To be fair, both Florida and New York have the same margin of error as California.

[skywaterblue]

It's quite funny to watch paleontologists speculate about dinosaur intelligence. When I last left off, they had downgraded the intellegence they thought that the Dromeosauride species possessed. (Probably in reaction to the Jurassic Park films.) Since then, however, there have been a bunch of studies showing that our current Corvus are quite bright and it's very likely that all the Dromeosauride family members had bird brains in addition to the feathers...

But then again, I'm a person who thinks elephants and orcas are likely sentient as well as apes. We're a lonely species, looking to the stars when our companions are all around us.

[jordan179]

But then again, I'm a person who thinks elephants and orcas are likely sentient as well as apes. We're a lonely species, looking to the stars when our companions are all around us.

Great apes, possibly some other primates, elephants, orcas, possibly other ceteceans, at least some parrots, at least some corvids ... and we're not acknowledging it, instead of responding with wonder and joy that we are not alone, we are twisting the evidence to tell ourselves that they are not sapient simply because (1) they probably aren't as smart as us (but why do other sapients have to be Wise Elders?), and (2) they have only very limited technology (but is being sapient really about having invented guns?). And of course the real reason (3) we want to feel superior and special (isn't being the only Earthlife that's invented rocketships and atomic bombs special enough

[skywaterblue]

If only someone took the idea of talking to them seriously. My first guess would be to try with orcas, since at least with whales we know they have names and dialects...

[jordan179]

If only someone took the idea of talking to them seriously.

Some people do, mostly behavioral scientists who have worked closely with great apes or parrots. Unfortunately a lot of them are considered to be flakes, even when they aren't. One problem is that you have to "personalize" or "anthromorphize" a sapient animal to communicate effectively with it: if you treat it as a "thing" it will notice this and become uncooperative. Penny Patterson's breakthrough with Koko came when she stopped treating her as just an experimental subject, and a lot of the scientific community feels that the obvious love Penny feels for Koko renders her results unreliable