Slumdog Smarts

Slumdog Millionaire makes a good point about intelligence: the chaiwallah knows a lot of answers, if the questions relate to his life experiences.

This thought sits on a serious problem with IQ testing, or with most standardized tests like the SAT, GRE or CAT. They don't test native intelligence. They test familiarity with a specific way of thinking, which is embedded in culture.

This is the reason why cultural minorities tend to do poorly on standardized tests. A favourite example from this piece by Malcolm Gladwell illustrates this point: Kpelle tribals from Liberia naturally group knives with potatoes, because knives are used to cut potatoes. While this is quite logical, standardized tests generally expect knives to be grouped with other tools, and potates with other root vegetables.

Unfortunately, this unsurprising and well understood limitation of standardized testing has led to horribly complicated racial profiling for university admissions in the USA, and in explicit, even more divisive, quotas in India.

Surely the more creative route is in designing culture-neutral tests? And in validating these instruments sufficiently to bring them into mainstream use?

This blog is idealistic enough to believe that better technology can at least alleviate really knotty political problems.

Seeing the ball like a football

Cricket fans know that a batsman who has spent a lot of time at the crease is hard to dismiss, because he is “seeing the ball like a football”. A batsman who is new to the crease is always easier to dismiss. He struggles to sight the ball.

This is true regardless of the quality of the light. A batsman who is in can bat on comfortably through the gathering caliginosity, while a new man at the crease struggles to sight the ball even in glorious sunshine. This has always been true, something cricketers accept as natural.

The mechanism that makes this natural just became apparent me, from this article by Atul Gawande.

Gawande’s piece is about an emerging scientific understanding about the nature of perception. The new realization: perception is mostly memory. The inputs coming in from the senses are thin/ low fidelity/ low resolution/ highly pixellated compared to the richness with which the brain experiences the sensory input. The mind fills in the blanks.

Our centuries long assumption has been that the sensory data we receive from our eyes, ears, nose, fingers and so on contain all the information we need for perception…Yet, as scientists set about analyzing the sensory signals, they found them to be radically impoverished…The mind fills in most of the picture...Richard Gregory, a British neuropsychologist, estimates that visual perception is more than ninety percent memory and less than ten percent sensory nerve signals…

Gawande’s article talks a lot about phantom limbs, and intense itches felt on injured tissues which have no nerve endings. These extreme examples are useful because they make a powerful argument; perception of a phantom limb can’t be determined by objective sensory experience, because there is no sensory experience. But to me, this theory is more interesting because of the light it sheds on everyday experiences.

A batsman who is in is literally seeing the ball better than a batsman who has just come to the wicket. His memory has more readily accessible images of the moving ball. He is therefore better able to make meaning of the sketchy data that his eyes pick up.

This is the reason it is hard to listen to an unfamiliar genre of music. The mind simply doesn’t have enough stuff in memory to fill in the blanks and enrich the music.

This is the reason it is hard to drive on unfamiliar roads. The driver literally sees less of the road. The eyes pick up the same volume of information as on a familiar road. But the mind doesn’t have a stock of memories with which to enrich the image.

This is the reason I enjoy watching cricket on TV more than I enjoy watching football. My mind has a bigger bank of cricket memories to draw on, simply because I have watched more cricket over the years.

There is an elaborate academic literature on how Caucasian-Americans are not very good at recognizing Blacks, and to a slightly lesser extent, how Blacks are not very good at recognizing whites. This has sometimes been interpreted as racism, but sheer lack of familiarity seems a simpler and less incendiary explanation. Interestingly, the effects are smaller in racially integrated schools and among children who live in integrated neighbourhoods.

This might also be the reason for the cognitive biases that Greg Pye's blog (and Kahneman and Taversky), keep talking about. The confirmatory bias happens because people, literally, don’t see evidence which goes against their prior beliefs without making a pretty substantial effort.

Attack of the Asian female clones

Glorious giants of the Appalachians are being killed off by insignificant-looking Asian females. And this has nothing to do with outsourcing, job losses, small towns, bitterness, guns or religion.

The Eastern Hemlock, a glorious native American tree that grows to a stature of 100m, the Sequoia of the Appalachians, is being wiped out by a tiny parasite, an aphid called the Woolly Adelgid. This has been observed and mourned in the New Yorker (I found the story thumbing through a back issue), the New York Times (way back in 1991) and in various pamphlets accessible with a Google search. However, what doesn't seem to have attracted comment is that the attacker is fatally flawed.

The aphids first came to America on decorative Japanese trees which were planted at Maymont, a public park in Richmond, VA. All the male aphids died. They feed exclusively on spruce sap and the males could not digest American spruce. A few females survived. Sans males, they had to reproduce by cloning. So, the threat to the Eastern Hemlock comes from clones of a very small number of individual female aphids. The clones were fantastically successful because they could colonize the Eastern Hemlock. As a predator gets established in America, or worst-case, as the Hemlock populations in the wild die out, the clones will also die. Clones are evolutionary dead-ends.

Should conservationists freeze Hemlock gene-plasm to re-populate the Applachians once the Woolly Adenids clones inevitably die? Makes sense. Just be sure to freeze a diverse pool of Hemlock gene-plasm. And establish an Eastern Hemlock worshipping cult whose rituals will remind initiates to perform this sacred task when evolution plays out and the clones finally die.

Science or religious war?

Are neo-classical and behavioural economics truly irreconcilable? Are the foundations of one of our great sciences crumbling before our eyes? Or, is behavioural economics a useful elaboration on the neo-classical tradition, that enriches rather than wrecks the great neo-classical insights?

My chips are on the enriching rather than wrecking side of the argument. Right here, right now, I clearly am in the minority. The sloppy-thinking chat tends to pit one tradition against the other.

Here is Elizabeth Kolbert, writing in the New Yorker.

"Rational calculators are supposed to consider their options, then pick the one that maximizes the benefit to them. Yet actual economic life, as opposed to the theoretical version, is full of miscalculations, from the gallon jar of mayonnaise purchased at spectacular savings to the billions of dollars Americans will spend this year to service their credit-card debt. The real mystery, it could be argued, isn’t why we make so many poor economic choices but why we persist in accepting economic theory."

Notice the condescension. "The real mystery...why we persist in accepting economic theory." She illustrates this self evident truth - that the assumption of rational economic behaviour is so broken that it is pitiful - with a personal anecdote.

Ms Kolbert was buying a book on Amazon for her work when a message popped up on her screen "add $7.00 to your order to qualify for FREE Shipping". She hesitated. Her nine-year-old twins wanted a Tintin book. She clicked it into the shopping cart and checked out, saving The New Yorker $3.99 and spending $12.91 of her own money.

Then comes the sucker-punch. "From the perspective of neoclassical economics, self-punishing decisions (such as this purchase of a Tintin comic) are difficult to explain." Really?

Neo-classical economics is very comfortable thinking about the Tintin comic as a bundle of information (about prices), services (delivery) and physical product (the comic book itself). Remember complements and substitutes from Econ 101? Amazon reduced the price of that bundle by flashing a free shipping advert on Ms Kolbert's screen, which reduced her tacit search costs, and by waiving the delivery charges. This prompted a purchase which otherwise wouldn't have happened. Lower price, more sales. This is the demand curve from Econ 101. There is no more unsurprising result in neo-classical economics.

Notice that the neo-classical theory makes no claims about the mechanism by which people maximize utility. In the neo-classical view people maximize utility the way a dog catches a frisbee. The calculations a dog needs to make to catch a frisbee are way beyond the scope of what a dog, or human, can consciously perform. Yet, dogs and humans successfully catch millions of frisbees every day.

This is where behavioural economics comes in. It sheds more light on the heuristics that people actually use while maximizing utility. This might prompt governments or businesses to use those heuristic mechanisms. For example, tax collections in America increased when taxes were collected at source. Once the money is in my bank account I am much more reluctant to give it to the government. Brilliant behavioural insight. By Milton Friedman, the godfather of neo-classical economics.

None of this challenges the one thought central to all of neo-classical economics: people respond to incentives. The way in which different people respond to different incentives is infinitely varied, which is why economic life is fascinating to observe.

My Family Owns a House in Telluride

Sometimes, things come together. Time, place, people, thought, soft afternoon sunshine, salty sea spray - everything - everything comes together in an exhilarating, intoxicating rush of adrenaline, testosterone, music and laughter that sculpts the soul, makes life real, and makes you who you are.

But things come apart again. Life moves on. The clocks didn’t all stop. We didn’t die then and there. The grey world that dawns the morning after Camelot, packing your bags for Faisalabad after winning the Ashes, it’s the hardest part. You still gotta do what you gotta do.

This haunting story is about the end of magic, about when the universe no longer has a center. Fortunately, the story still is on the New Yorker’s website.

http://www.newyorker.com/fiction/features/2007/11/19/071119fi_fiction_nelson

Tintin

Found a superb article on Tintin in a back issue of the New Yorker, by Anthony Lane. The highlights:

- General de Gaulle declared that Tintin was his only international rival. He was envious, perhaps, of not just of Tintin's fame but of his defiantly positive attitude. Both figures can be recognized by silhouette alone

- Stimulation in Herge's solid, Catholic, bourgeois youth in Belgium came from his exploits as a scout. There remain in Tintin traces of the try-anything, do-gooding spirit of the scout troop

- Tintin was first serialized in 1925 in a daily newspaper which was described in it's masthead as a "Catholic and National Newspaper of Doctrine and Information". The editor had a framed picture of Mussolini on his desk

- Tintin in Congo (1931) is an unmitigated parade of racial prejudice, with bug-eyed natives swaying between ignorance and laziness. Herge later redrew the comic...and claimed that his concept of the Congo was no different from that of his compatriots at the time

- A crucial happening was Herge's encounter with Chang Chong-chen, a student at the Brussels Academy of Fine Arts. Chang suggested using real events as the inspiration for his adventures...the ouevre would have a historical value...The Blue Lotus (1936) was born

- Chang returned to China and lost touch with Herge. Tintin in Tibet (1960) showed Tintin searching for a lost friend called Chang. In 1981, two years before Herge died, the real Chang came to Belgium for a tearful reunion. Chang had shown Herge how to beat back prejudice: just tell the truth

- Herge's principle: the further your reach, the more compelling your duty to get it right. Herge knew he would not be able to emabrk on Tintin in Tibet without amassing photographs of monastries, lamas and chortens, all of which would be copied in fanatical detail in his book. I was thrilled to read Hindi in Tintin which was real

- Belgium was occupied. Herge - like Wodehouse, who was interred as an enemy alien - not only survied but bloomed into one of his most flourishing periods. Once the war ended, both were interrogated about the nature and intensity of their collaboration. Both pleaded guilty of innocence; neither ever dispelled the shadow of suspicion. Herge's (and Wodehouse's) ability to avert his gaze from evil verges on the chronic

- Commentators are both enticed and exasperated by how little he gives away. In particular, the Tintin who gazes out from the cover of The Castafoire Emerald (1963) shushing the reader with a finger held to his lips

- There is no sex. Tintin passes increasing portions of his time with an unmarried seaman, yet it seldom occurs to us to question their rapport. He never has a girlfriend, and never expresses the need for one. He has no parents or siblings. We are unsure whether he counts as a child himself. He reminds me, if anyone, of Charlie Brown. Enid Blyton, and maybe even JRR Tolkien, would be at home in this beautiful, adventure-filled, but asexual world

- Tintin may be too constrained for American tastes, being posessed of no superpowers. He is Clark Kent without the phone booth, although Clark at least had a paying job , wheras Tintin, nominally a reporter, never receives a paycheck or files a story

The Little Mermaid

Some gems from opera director Francesca Zambello's interpretation of Little Mermaid for a Broadway production:

- The mermaid ascends to the surface of the sea, her tail unfurling to reveal shapely legs. There is so much metaphor in that. It is like a rite of passage, her first menstrual cycle

- The wish-fulfillment element would give it broad appeal. Show me anybody in the world who hasn't wanted to be someone else. That's a universal theme. Everybody sees themselves as an outsider

- It was possible to interpret the little mermaid flight from the confines of the sea as a gay theme. The reality is that there are only two minorities who are born into families: disabled people and gay people. Every other minority is born of a family. That Ariel is an outsider in her own family connects...

- Mermaids have no genitalia. That's something you don't really think about until you work on mermaids, but then you think about it a lot.

Swingers

Been working through a New Yorker article about bonobos in the wild. Was reflective, well researched, emotionally rich. Loved it. Click through to read

http://www.newyorker.com/reporting/2007/07/30/070730fa_fact_parker

It turns out the popular belief that bonobos are "...into peace and love and harmony...equal parts dolphin, Dalai Lama, and Warren Beatty..." is at best a partial truth. This belief was based largely on observing a group of pubescent bonobos in captivity. They outgrow their rampant teenage sexuality. They have their own systems of social dominance. They are capable of murder, maybe even the murder of infants.

A trained social psychologist wouldn't be surprised. Their experiments show that most human behaviour is situational. Almost any human being can be kind, cruel, nurturing, willfully destructive, cynical, starry-eyed, social, organized, lazy...anything...in the right context and with the right conditioning. Other apes are just like us.

An interesting side-story I picked up was that we've been fooling ourselves about the nature of great apes for over a long time. Consider:

- "The bonobo of the modern popular imagination has something of the quality of a pre-scientific great ape, from the era before live specimens were widely known in Europe. An Englishman of the early eighteenth century would have had no argument with the thought of an upright ape, passing silent judgment on mankind, and driven by an uncontrolled libido."

- "(In 1972...Goodall had confidence that chimpanzees were “by and large, rather ‘nicer’ than us....In 1974...Goodall witnessed...the War in Gombe. A chimpanzee population split into two...one group wiped out the other, in gory episodes of territorial attack and cannibalism".

Consider the main theme of the article: bonobos can be from Mars. Consider the Margaret Mead mythology of idyllic primitive societies (I learnt that this was just flatly untrue from Jared Diamond's Guns, Germs and Steel).

There is a pattern here. We seem to have a need in our collective unconscious for a noble nearly-human figure. Letting that need project on to some poor unsuspecting great ape or primitive society leads to really bad science. Fortunately, the great apes haven't physically suffered because of that bad science.