Grasping Reality with Both Hands: The Semi-Daily Journal Economist Brad DeLong

Hoisted from Comments: The Dawn of Humanity

Hoisted from Comments: The Dawn of Humanity

Grasping Reality with Both Hands: Brad DeLong's Semi-Daily Journal: The Dawn of Humanity: What astonishes me is the speed. They've got the origin date at -56,000, and the oldest modern human remains in Australia are -40,000. The route from East Africa across Asia to Northern Australia is 10K+ miles, which means humans were expanding at close to a mile a year. That's just unbelievably fast.

We have all sorts of branches of homo surviving stably for a million plus years all over africa, asia, and europe, and this new branch comes out of Africa and by the end of the Great Migration, only a little over ten thousand years later, they are building boats to sail to Australia. And wiping out or out-competing every one of our homo sibling species on the way.

The Singularity is truly in our past.

Posted by: tavella | January 23, 2007 at 05:15 PM

Strong AI

Run across anybody lately who says that we will never be able to build a human-level intelligence and fit it into a breadbox?

dd-b over at Tor has the answer:

Tor.com / Science fiction and fantasy / Blog posts / The Singularity Problem and Non-Problem: I'm always surprised to find people who "doubt" strong AI.... [H]uman beings are themselves examples of strong AI. I find it amusing to hear people arguing that they cannot, in fact, exist...

Of course, I think it will be much cheaper to make adult human-level AI entities via a twenty-year production process using unskilled labor--at least half of all possible two human teams can do so!--then employing skilled computer scientists. But what do I know?

Higher-Level Languages and Genetic Programming

The Quintessence writes::

Quintessence of Dust: Wait...did you say "eldritch?": Comparison of the various chordate genomes reveals that there are very few chordate-specific genes. Specifically, the authors described 239 "chordate gene novelties" out of 22,000 genes in the lancelet. The nature and function of these genes is intensely interesting, and indeed the authors devote a separate report to issues related to this. But think about it: only 1% of the genes in chordates (vertebrates and all their relatives) are "novel" among genes from all other organisms. So if the toolbox isn't all that different between lancelets and lions, despite divergence at least 550 million years ago, then what is different? Anything? As John Timmer notes on Nobel Intent, the authors could find relatively few examples of regulatory DNA sequences that are conserved between lancelets and vertebrates, pointing to the likelihood that changes in regulation of a (mostly) common genetic toolkit is a major factor in evolution of form. (Okay, so that was just a plug for evo-devo. It's my blog.)...

Now that is scary. The DNA genome is best conceptualized not just as machine language for the cell and the organism, not just machine plus assembly language, not just machine plus assembly language plus Fortran, but all of those and overlaid over the whole, controlling everything, the highest-level genetic code for our humanity written in the molecular equivalent of Java...

Daniel Davies, Diet Coach

He writes:

Daniel Davies: Fat Hominid: There’s a paper to be written at some point on the economics of fad diets... a rich source for the self-organising systems literature and a good case study of how irrational and somewhat self-destructive beliefs spread through proselytisation.... [N]early everyone’s digestive system is different.... Different foods agree and disagree with different people.... [F]ad diets... can... be modelled as more or less spanning the possible combinations of foods.... [E]very now and then, someone is going to come across a fad diet which really really really works, for them, because it happens to not include whatever food is giving them their current digestive troubles.

Someone like that is very likely to become an evangelist for their preferred fad diet; after all, they have first-hand empirical evidence that it really really really works. And sudden relief from digestive discomfort, or very rapid weight loss, is an experience the emotional impact and profundity of which should not be underestimated....

Of course, the vast majority of people on fad diets are getting no real benefit from them, other than from the incidental factor that most of them are basically calorie controlled (either by design or, per Atkins Diet, de facto by simply being such inconvenient and unpleasant ways to eat). Thinking about these sorts of things and their spread through the community gets you onto the subject quite quickly of Charles Mackay and Extraordinary Popular Delusions, which is why it’s a bit of a disappointment to me to see that a sharp cookie like Nassim Nicholas Taleb appears to have fallen hook line and sinker for a fad diet...

P.Z. Myers on the Platypus

P.Z. Myers says that the platypus is a fine animal--it has just been pushed around by different evolutionary pressures over the past 160 million years than have modern mammals, reptiles, and birds. I say that the platypus is an otter gone horribly wrong:

Pharyngula: The platypus genome: Every organism is going to be a mix of conserved, primitive characters and evolutionary novelties — a mouse is just as "weird" as a platypus from an evolutionary perspective, since each is the product of processes that promote divergence from a common ancestor, and each are equidistant from that ancestor.... [M]odern echidnas, elephants, and emus are all products of different evolutionary trajectories through history, and no one by itself is a representative of the ancestral condition. We derive the ancestral state by comparison of multiple lineages... [this] adds another lineage to the [genomic] data set, one that diverged from ours over 160 million years ago. It is a lens that helps us see what novelties arose in that 160 million year window....

So what are the details that we've learned from the platypus? One important message is the unity of life. The platypus has about 18,000 genes; humans have 18-20,000 genes. Roughly 82% of the platypus genes are shared between monotremes, marsupials, eutherians, birds, and reptiles....

An interesting specialization in the platypus is the evolution of venoms. The platypus has small, sharp spurs on its hindlimbs that it uses to inject defensive poisons into predators, a very unusual feature not found in other mammals. Where did these venoms come from? As it turns out, by duplication of genes that have other functions, with subsequent divergence, and many of these genes also come from the innate immune system. In particular, there are a set of proteins called the β-defensins... the bullets of the immune system; they can bind to viral coat proteins, they can punch holes in bacterial membranes.... The platypus has repurposed these genes, making copies that have been selected for more effective toxicity when injected into other animals.

One very cool observation is that these are also the same proteins used in venomous reptiles... two distant relatives, the lepidosaurs and the monotremes, all use β-defensin derived venoms. Does this imply that their last common ancestor also used these venoms? No, and this is where the details are important. Venomous snakes and the platypus have different duplications of the β-defensin genes... these are independently derived features, not primitive at all... convergent evolution....

One virtue of the platypus is that it provides a relatively closely related outgroup to help tie together, and give perspective on, the various mammalian genome projects. It's all part of the big picture in defining what a mammal is...

P. Z. Myers Has a Quibble with the Excellent "Inner Life of a Cell" Video

It has aroused his ire. From P.Z. Myers:

Pharyngula: Buffeted by the winds of chance: why a cell is like a casino: This is what bugs me about the "Inner Life of a Cell" video. They're portraying the behavior of single molecules, and the movie has them dancing a slow waltz, steadily moving through fixed patterns. It should look more like a mosh pit filled with meth addicts; real chemistry doesn't direct single molecules, it shifts overall equilibria so that the random activity of single cells has certain probabilities of throwing them off a thermodynamic cliff into a new state. Those kinesin "feet" [on the motor transporter molecule] should have been pitter-pattering all over the place, occasionally falling into a more stable position that led them in a particular direction -- what was needed was a portrayal of a hyperkinetic ratchet. Anyone who uses a microscope or looks at the activity of small numbers of molecules or studies thermodynamics and reaction equilibria (like, say, a chemist or biochemist) ought to be familiar with the stochastic properties of the world on such a small scale.

The closest example on a macro scale that I can think of is a casino. People go in and out of a casino, and they engage in many small probabilistic events. Some people win big, some lose big, and all states in between are represented... but the house always has its small, advantageous odds in its favor. They don't shake down the crowd deterministically and demand a cut from each, instead what they have done is basically tipped the the reaction equilibrium gently to favor the transfer from one state -- your pocket -- to another state -- their bank. From the aggregate kinetics of a great many transactions with only a tiny edge in one way, they have constructed a powerful and reliable siphon hose to draw off your money. (I suspect that's also one, perhaps unconscious, reason why gambling establishments are fanatical about keeping out people with even a hint of a successful gambling system; it doesn't take much of a shift in the percentages to reverse the flow in their money siphon.)

If you watch single individuals in the casino, rather than the rising total profits for the whole institution, you wouldn't see a smooth and steady drain of money, though. Over long periods of time the trend would appear, but moment-by-moment? No. You'd see a herky-jerky random pattern of wins and losses.

And that's what I would want to see portrayed in my ideal version of an animation of the chemistry inside a cell: not a ballet, a jostling mob on an uneven floor. Show me more noise and chaos.

Robert Waldmann Has a Big Problem with the Anti-Prozac Meta-Analysis Study

Robert Waldmann has a big problem with and talks back to the anti-Prozac meta-analysis study of Hirsch et al.*.

Mark Liberman summarizes the meta-analysis data in a nice picture:

The x's are studies. The vertical axis shows the improvement in mood for people being given the placebo--the sugar pill. The horizontal axis shows the improvement in mood for people being given the antidepressant, both according to the Hamilton Scale of Depression.

People being given the placebo improved their mood a lot--by 7.8 points, which is a relatively big deal on the Hamilton Scale: feeling that you are taking control over your Depression by getting involved in a cutting-edge medical study, the fact that a group of research scientists are paying attention to you, and the passage of time together do a lot of good. But the people being given the actual anti-depressants improved their mood by even more. Let's turn the mike over to Hirsch et al.:

[W]eighted mean improvement was 9.60 points on the HRSD in the drug groups and 7.80 in the placebo groups, yielding a mean drug-placebo difference of 1.80 on [Hamilton] improvement scores.... [which] easily attained statistical significance [at the 0.001 level, in fact--much better than the 0.05 level]...

Subjects given Prozac improved their mood by an extra 1.8 points on the Hamilton scale. This difference is not due to chance sampling error--it is, statistically, very significant. The pills are really cheap to make. There is an upside. Better Living Through Chemistry.

So what's the problem with Prozac? The problem, according to Hirsch et al., is that the difference of 1.8 points on the Hamilton Scale:

does not meet the three-point drug–placebo criterion for clinical significance used by NICE [Britain's National Institute for Health and Clinical Excellence]...

Where does this requirement that no therapy for Depression is worthwhile unless it improves the Hamilton Scale score by three points come from? The weblog "Pyjamas in Bananas" finds a quote:

Pyjamas in Bananas: No research evidence or consensus is available about what constitutes a clinically meaningful difference in Hamilton scores, but it seems unlikely that a difference of less than 2 points could be considered meaningful. NICE required a difference of at least 3 points as the criterion for clinical importance but gave no justification for this figure...

Who wrote this? Irving Kirsch, lead author on the anti-Prozac study.

And it is at this point that the economist in me wants to reach for his revolver. A declaration that a real-world solid statistically-significant improvement in people's quality of life is not "clinically significant" is inadmissable unless it is motivated by a proper analysis of opportunity costs: a conclusion that the resources devoted to this therapy would have a higher value and a better alternative use in some other therapy. It cannot rest on an arbitary number that some organization pulls out of its a--.

Even worse, Robert Waldmann points out, is that the Guardian's health editor Sarah Boseley doesn't understand the article she is reporting on:

Prozac, used by 40m people, does not work say scientists: Analysis of unseen trials and other data concludes it is no better than placebo: Prozac, the bestselling antidepressant taken by 40 million people worldwide, does not work and nor do similar drugs in the same class, according to a major review released today.... When all the data was pulled together, it appeared that patients had improved - but those on placebo improved just as much as those on the drugs...

Waldmann comments that Boseley is:

totally dishonest, totally innumerate or both. 1.8 > 0. Patients on Placebo did not improve just as much as patients on SSRI's... this isn't even a case of treating a statistically insignificant difference... as... proof that the true value is zero.... "Irving Kirsch, Brett J. Deacon, Tania B. Huedo-Medina, Alan Scoboria, Thomas J. Moore & Blair T. Johnson" find a significant additional benefit of taking a SSRI rejecting the null of no benefit with a p value of "<0.001"... overwhelmingly strong evidence that SSRI's cause improvement in depression.... Oddly big Pharma, which spends huge amounts of money on advertising, doesn't seem to have managed to hire anyone intelligent enough to point out that 1.8 > 0...

A Leukocyte Goes About Its Business...

Ah. Here this is with its narration:

In some ways, I think it is better without the narration--without the dry description of what is going on. It is in a sense more marvelous when it is incomprehensible, or rather uncomprehended.

See also:

http://delong.typepad.com/sdj/2006/10/yet_more_on_ins.html

http://delong.typepad.com/sdj/2006/10/biologists_weep.html

http://sparkleberrysprings.com/v-web/b2/index.php?p=737&c=1

http://cosmicvariance.com/2006/09/29/the-cell-is-like-tron/

http://www.studiodaily.com/main/searchlist/6850.html

Melanin Level Blogging

From National Geogrqphic:

Skin Online Extra: Melanin, the brown pigment in the skin, acts as a natural sunscreen. It protects against UV, and populations in the tropics are darker skinned since there is more sunlight where they live. UV ages the skin, causes skin cancer, and--most significant to Jablonski and Chaplin's work—-breaks down folate, essential vitamin B needed for cell division and producing new DNA. Pregnant women in particular require large amounts of folate to support rapid cell division in the embryo.... So if a higher melanin level is so beneficial, why isn't everyone dark-skinned?

Jablonski and Chaplin concluded that modern humans... evolved in the tropics, where they were exposed to high UV levels. But... away from the equator, where UV levels are lower, humans became fairer so as to allow enough UV radiation to penetrate their skin and produce vitamin D, the "sunshine vitamin," also obtained from eating fish and marine mammals... essential for maintaining healthy blood levels of calcium and phosphorous, and thus promoting bone growth. Skin color... becomes a balancing act between the evolutionary demands of photo-protection and the need to create vitamin D in the skin.

But things aren't always what they ought to be. That is the case with Eskimos and other inhabitants of northern Alaska and northern Canada. "Looking at Alaska, one would think that the native people should be pale as ghosts," Jablonski says. One of the reasons they're not is that these populations have not lived in the region very long in terms of geological time. But more importantly, their traditional diet is rich in fish and other seafood.... "What's really interesting is that if these people don't eat their aboriginal diets of fish and marine mammals, they suffer tremendously high rates of vitamin D-deficiency diseases such as rickets in children and osteoporosis in adults," Jablonski says...

Interesting that the Eskimos have not lived long enough near the North Pole for their skin color to evolve...

Note to Self: Brainstorming: A Health Care Symposium?

What would I most like to see in a quick health care symposium this fall/winter for The Economists' Voice?

(A) Ask representatives of the three major Democratic campaigns to comment on the excellences of the other campaigns' plans:

• Obama rep. on the excellences of Clinton and Edwards
• Clinton rep. on the excellences of Edwards and Obama
• Edwards rep. on the excellences of Obama and Clinton

(B) Ask people who know the legislative politics to explain:

• What significant improvements are impossible because the pharmaceutical lobbyists will block them?
• What significant improvements are impossible because the insurance lobbyists will block them?
• What significant improvements are impossible because the doctor lobbyists will to block them?

(C) Taking a broader view:

• What's wrong with all three plans: the view from the right?
• What's wrong with all three plans: the view from the left?

(D) Taking a broadest view:

• What public-policy mediated changes in health care financing, delivery, education, and information would produce the biggest improvements in Americans' health status, and why?

Squirrels Wage Fifth-Generation Information Warfare Against Rattlesnakes!

Annals of science:

Squirrels wield a hot, secret weapon - life - 13 August 2007 - New Scientist: Jeff Hecht: Squirrel waves hot tail at rattle snake--Watch the full-size video; Squirrel waves cold tail at gopher snake--Watch the full-size video: It's Californian ground squirrel versus rattlesnake in a potentially lethal showdown. But the squirrel has a secret weapon that until now has remained invisible to the human eye. The ground squirrel heats up its tail then waves it in the snake's face - a form of harassment that confuses the rattler, which has an infrared sensing organ for detecting small mammals. This defensive tactic remained invisible to biologists until they looked at the animals through an infrared video camera. Now they believe that many other animals might be using infrared weaponry to ward off potential predators. Young California ground squirrels (Spermophilus beecheyi) are easy prey for snakes, so protective adults harass the predators while puffing up their tails and wagging them.

Graduate student Aaron Rundus and his supervisor Donald Owings of the University of California, Davis, wondered how this might affect the snakes’ interaction with the adult squirrels. So he borrowed a $35,000 infrared camera from another scientist and spied on squirrel-snake stand-offs. He saw the adults’ tails heat up, presumably due to increased blood flow, when they were warning rattlers away – making the squirrel appear larger to the snake’s infrared organ. Confronted with a gopher snake, which has no infrared sensory organ, the squirrels wagged their tails but didn’t bother to warm them up first.

Tests with robotic squirrels confirmed that a warmed squirrel tail made rattlesnakes more likely to act defensively, say Rundus and Owings.

The squirrels themselves do not see in infrared, so they cannot see another squirrel's tail heating up. But the snakes can, proving that the squirrels have evolved a specific way to deter rattlesnakes. “It taught us to focus on the perceptual world of the animal we’re studying” rather than thinking only of human perceptions, says Rundus.

Perhaps the scariest thing here is that UC Davis has developed "robotic squirrels"--I assume as part of a dual-purpose research project.

Pharyngula: Journamalism at Wired

Pharyngula watches journamalism as practiced at Wired. Memo to Chris Anderson: you really do need to do better than this:

Pharyngula: Someday, Cosmopolitan will ask me to write a piece on beauty tips, too: My opinion of Wired magazine just dropped a couple of notches. They've got Gregg Easterbrook pontificating on a science issue, the origin of life. Easterbrook is a sports writer with absolutely no clue about science--I've commented on his incompetence a few times before (OK, more than a few times). This time he's soberly stating that no one has done any research on abiogenesis since Miller/Urey, or what they've done is a series of failed experiments, and that there are no hints in nature about the chemical origins of life, therefore, maybe a god did it --while completely oblivious to the fact that no one has ever done any research on gods or higher beings, and that there is no evidence for their existence. The man is an idiot. I am still utterly baffled why anyone consults that twit for his opinion on science.

Being Omnivorous

We turn left coming out of the driveway, and we are immediately confronted with: quail, seventeen adult quail, and deer, two adult does. The quail do not quail at the car: instead, they run aimlessly in rapid circles as the car approaches, and then take wing with a whir. The deer stare at us, as if wondering whether or not to approach the car for food. And then they amble off toward the lawns. In the distance there is the gobbling of turkeys--refugees seeking sanctuary from the retirement community of Rossmoor, where sharpshooters are hunting them with silenced rifles. My wife tells of once being in the car and running into a male turkey on the driveway, which looked at her and spread its tail feathers in a testosterone-crazed dominance display--thinking that it could drive off a Subaru, and thus preserve his exclusive sexual access to the hens. Truly a bird of very little brain...

Ever since my wife learned that I was on a committee with Michael Pollan, author of the truly excellent if slightly Berkeley twee The Omnivore's Dilemma, she has been pressing me to invite him to dinner. I have resisted, being scared that she would greet him at the door with a net, and say: "We're having quail this evening. Would you please catch us a dozen? They're under the blackberry bushes" or "Here's the dried corn, the mortar, and the pestle: would you please make us some masa?" or even worse, "Would you please evolve us some maize via selective breeding from this teosinte plant?"

But do go read The Omnivore's Dilemma. It is truly excellent.