« Like CPUs and Operating Systems, Countries Matters | HomePage | On Teaching and Learning »

Monday, September 04, 20061157388600

Genetic and Experiential Individuality

"Fruity Genes," by Michael Morgan, The Guardian, 22 March 2001, http://www.guardian.co.uk/Archive/Article/0,4273,4156478,00.html.

J.R. of Edgewise asked me for comment on Rudiger Gamm, a mathematics prodigy. I've commented on genetics and expertise before, so Edgewise asked me if perhaps this Mr. Gamm shows that learning is more important than "evoking" in creativity. After all, Rudiger's talent developed only in adulthood and through purposeful practice.

We know that not all genetic factors are present at birth. Political orientation, for instance, seems to be evoked only after our twentieth birthday. More trivially, breasts are absent in newborns but become very prominent later in life. Likewise, we know that intelligence within a culture is highly heritable (more than half of variation can be statistically explained by genes, some from shared environment, and the rest from unique experiences).

In fact, more can be explained by unique experiences than shared experiences. We are wired for struggle, for real mentoring, for victory. Humans learn best when individuals have a stake in the outcome of education, in the form of games, tournaments, and competition. Giving the same lecture or same home environment to two children means very little. Giving them meaningful unique experiences means a lot.

Intelligence, properly understood, is not a story of genes v. individuality. It is a saga of genetic individuality and experiential individuality against the authoritarian, bureaucratic, inhuman sameness of the modern world.

When I see research showing how intelligence has strong genetic factors, I say "all right!" When I see research that we can use purposeful practice to become experts, I saw "wonderful!" From the moment we are conceived, we are unique individuals. God did not make just one a Generic Man who varies based only on input. He has created billions of individuals who are called individually but have the freedom to do what they choose.

He gave every individual himself to control. And us a world to rule.


What is 99 raised to the fifth power? Most people could only reply like the computer Deep Thought: "It's going to take a little time." Rudiger Gamm is a famous calculating prodigy from Germany who can solve such problems in his head. A recent paper in Nature Neuroscience has analysed just what goes on his head when he is doing so. Using the fact that calculation is to the brain what sex is to the penis, scientists observed blood flow to the parts of the brain that were most active during arithmetic exertion. They found that the expert used different parts of the brain from normal people.


The connection between brain anatomy and creativity is a site of interesting research. This extends beyond expertise in mathematics to much more commonplace things, like expertise in sensation in your little finger. If one of your fingers is cut off, the other fingers will "take over" the portion of the brain that formerly was dedicated to it. Likewise, autistics will use portions of the brain normally dedicated to understanding rule-based systems to process social interactions. (The reverse is also probably true: bimbos would use portions of their brain dedicated to social interaction to process rule-based systems. Thus the origin of every blonde joke in the world...)

Psychologists say that the Victorian phrenologists were wrong to measure people's abilities by feeling bumps on the skull. But phrenologists were right in thinking that different parts of the brain carry out different tasks. The specialisation can be quite astonishing. The brain responds to the sight of fear in another person by increased activity in an almond-shaped structure called the amygdala. The pulvinar region is involved in directing our attention ( pulvinus is Latin for a couch, so presumably it is also involved in watching television).


The most famous special-purpose region is Broca's Area, which is probably the seat of Chomsky's Universal Grammar. (Interestingly, Broca's Area appears to be older than the portions of the throat dedicated to talking. Make of this what you will...)

Another article in Nature Neuroscience describes different areas of the brain responding to "monetary rewards and punishments." When someone gives us money one part of the brain expresses our delight, while if they take it away, another part of the brain indicates our displeasure.


And, more substantively, will reward us if we see cheaters punished. Male brains reward them more for observing Justice then female brains, however.

Chris Tarrant, please note. More astonishing still, is the claim that a highly specific part of the temporal lobe is activated when people are naming vegetables: could this be the long-awaited answer to whether a tomato is a fruit or a vegetable?


Silly, but it would be useful to know whether in the Era of Evolutionary Adaption what the distinction was. How would one usefully categorize vegetables against fruits in mobile hunter-gatherer societies?

One's brain is powerfully activated by learning that there are only 30,000 genes in the human genome. That is only twice as many as in the fruit fly, not renowned for its calculating ability, and about as many as a garden weed. Geneticists are sorely vexed that so few genes are required to make something as complex and clever as a geneticist. Even if 15,000 extra genes made our brain, how many would be devoted to the naming of vegetables? At the Millennium Evening at the White House one scientist said: "If I gave you a parts list for the Boeing 777 and it has 100,000 parts, I don't think you could screw it together and you certainly wouldn't understand why it flew."


The paragraph ends well but begins very weak. Even if one considers a "gene" to be just a "word," it's easy to see how an infinite number of thoughts may be assembled from as little as 850 words." But a "gene" is not just passive. A gene can say something like "jump 2 forward," "jump five back," and even be part of logical expression "if ____, then jump 3 forward, otherwise, jump 5 back." Recent research indicates most genes may function as conditionals most of the time. This means that when genes are "read" in processes, the "genetic code" is just a computer program for generating the actual text that will be interpreted.

But what is the parts list for the brain? Would it specify each neurone and its connections with every other neurone (impossible, since there are 10 to the 14th power connections and you don't need to be a mathematical prodigy like Rudiger Gamm to see that this is more connections than genes)? Or is it a list of vegetables? Even the latter seems unlikely, because the number of genes needed to programme a set of several million neurones to recognise a vegetable would be enormous.


Similar error here. The author assumes that the genetic code is similar to a human-produced program where each gene has one task. This is how I would write a program, because it is easier to debug. However, our bodies are the production of evolutionary systems design, where each gene has an arbitrary number of tasks.

The difference between the human brain and a Boeing 777 is that the growth of the brain can be guided partly by experience. If the 777 emerged from its shed in Seattle as a bicycle, which got better and better at flying by accumulating more parts and flying hours, we might have a better analogy. What we know about the development of our brain is that it begins growing like the undisciplined first draft of a book. Huge numbers of cells and connections between cells are grown which have no function in the final product. The brain then obeys the wise advice of Dr Johnston: "Read over your compositions, and where ever you meet with a passage which you think is particularly fine, strike it out." The exuberant cells and connections are gradually weeded out until the sparse, adult structure is complete.


More or less. Most portions of the brain are "built" by trimming away from an existing mesh, others are constructed by building up from a smaller network. Some portions "float" to their preferred position through childhood, while others "anchor" to where they first emerge.

A case where "striking it out" is not done is the walking reflex. Move a newborn along the floor while lifting him by his shoulders, and his legs will kick in a reflexive attempt to walk. This ability appears to go extinct until some time later, when a newborn stars "really" learning to walk?

Or does it? Hold a one year old over the floor, and no movement. Hold him over an underwater treadmill, and there's the old reflex kicking in. Even though the "walking" module is only partially functional at birth, and is overwhelmed by a high fat-to-muscle ratio through early development, it stays operational. It ignores that law of neural networks: reward success, punish failure.

Meanwhile the brain is receiving input. Even in utero, the baby hears sounds, and since it receives no light, the retina invents virtual reality to send nerve impulses back to the brain. How important are these signals for development? We don't know, but it is a fair bet that natural selection has discovered a cunning plan to make a few genes stretch a long way. Our immune system can recognise literally millions of foreign proteins. It uses only 300 genes. You can make a lot of different permutations from 300, as football pools addicts know, to their cost.


Good analogy.

Rudiger Gamm has trained his prodigious arithmetical ability for several hours each day for years. Somehow he has altered the wiring diagram of his brain. We won't find the "parts list" for his achievement in the human genome.


This concludes the article. The important message is this: we are not biologically determined. We are not environmentally determined. Both matter.

Post a comment