What Is General Intelligence?
In response to one of my posts discussing IQ recently, a commenter contended that, for IQ to be truly measuring "general" intelligence, the test must be aimed at gauging abstract reasoning using symbols, otherwise what it measures would not really be general.
I think this is exactly backwards. Abstract reasoning using symbols is a very specialized form of intelligence, highly developed in only one species (as far as we know), and, even then, engaged in as a primary life activity by a very small percentage of the members of that species.
General intelligence is the intelligence that steers one around in day-to-day life, so that one avoids being hit by cars, doesn't eat poisonous things, stops arguing before making someone so angry that they beat or kill you, figures out how to get enough to eat, and so on. Think about it this way: You all know some Cal Tech or Harvard or MIT grad with a 4.0 average, about whom it is said, "He has trouble crossing the street by himself: no common sense whatsoever." This person probably scores 160 or 180 on an IQ test, because it is a test designed by nerds (like you and me!), and tests the sort of specialized intelligence we possess. What our hypothetical braniac lacks is general intelligence.
What do you think the IQ of the average NBA player is? Whatever it turns out to be, I doubt it is exceptionally high. Yet, these people have to continually make split second decisions about extremely complex and evolving situations with a multitude of relevant factors. And they are where they are not only because of their physical skills, but also because they are among the very best in the world at making those decisions. These are very intelligent people, whose intelligence just doesn't happen to be captured by an IQ test.
If you are hiring someone to train as a computer programmer, the results of an IQ test might be a pretty good predictor of success. And in a society that highly values such skills, it might be a pretty good predictor of income. But it ain't general intelligence.
I think this is exactly backwards. Abstract reasoning using symbols is a very specialized form of intelligence, highly developed in only one species (as far as we know), and, even then, engaged in as a primary life activity by a very small percentage of the members of that species.
General intelligence is the intelligence that steers one around in day-to-day life, so that one avoids being hit by cars, doesn't eat poisonous things, stops arguing before making someone so angry that they beat or kill you, figures out how to get enough to eat, and so on. Think about it this way: You all know some Cal Tech or Harvard or MIT grad with a 4.0 average, about whom it is said, "He has trouble crossing the street by himself: no common sense whatsoever." This person probably scores 160 or 180 on an IQ test, because it is a test designed by nerds (like you and me!), and tests the sort of specialized intelligence we possess. What our hypothetical braniac lacks is general intelligence.
What do you think the IQ of the average NBA player is? Whatever it turns out to be, I doubt it is exceptionally high. Yet, these people have to continually make split second decisions about extremely complex and evolving situations with a multitude of relevant factors. And they are where they are not only because of their physical skills, but also because they are among the very best in the world at making those decisions. These are very intelligent people, whose intelligence just doesn't happen to be captured by an IQ test.
If you are hiring someone to train as a computer programmer, the results of an IQ test might be a pretty good predictor of success. And in a society that highly values such skills, it might be a pretty good predictor of income. But it ain't general intelligence.
No, I don't know any grads from those institutions, and I still can't crack 140 with a SD of 15. I don't know that I resemble the term "nerd", but I certainly understand what you're trying to drive at.
ReplyDeleteThe simple fact is that intelligence tests will be highly subjective based upon what qualities that the writers of the test value. I ultimately have my own subjective test, and it basically rests on the question: can you learn new things effectively, and can you put them to practical use? Some of the people in my life who would most assuredly score very poorly on an IQ test also happen to be those who I find to be the most intelligent based upon my own criteria.
The speed of problem solving would serve as a reliable measure, assuming one could construct novel problems that didn't have the sort of biases that current I.Q. tests have.
ReplyDeleteI think people with high I.Q.s could display greater general intelligence if they learned to be present. The guy can barely cross the street by himself because he's using all his cpu cycles on something else.