Wednesday, November 13, 2013

More GIGO?

This one is a doozie.
Title: On the Nature and Nurture of Intelligence and Specific Cognitive Abilities: The More Heritable, the More Culture Dependent
Psychological Science OnlineFirst, published on October 8, 2013 as doi:10.1177/0956797613493292

Abstract

To further knowledge concerning the nature and nurture of intelligence, we scrutinized how heritability coefficients vary across specific cognitive abilities both theoretically and empirically. Data from 23 twin studies (combined N=7,852) showed that (a) in adult samples, culture-loaded subtests tend to demonstrate greater heritability coefficients than do culture-reduced subtests; and (b) in samples of both adults and children, a subtest’s proportion of variance shared with general intelligence is a function of its cultural load. These findings require an explanation because they
do not follow from mainstream theories of intelligence. The findings are consistent with our hypothesis that heritability coefficients differ across cognitive abilities as a result of differences in the contribution of genotype-environment covariance. The counterintuitive finding that the most heritable abilities are the most culture-dependent abilities sheds a new light on the long-standing nature-nurture debate of intelligence.

----
What this is saying is that the more the culturally loaded the test that the IQmetricians use, the more it appears to be heritable (i.e., a larger fraction of the observed variation in test scores is attributed to genetic variation).  Put that in your pipe and smoke it.   My personal unexpert opinion is that if this result holds then the whole field of IQology has been shown to be gibberish.

Let's reason it out. 

1. A test that say, consists of puzzles based on visual patterns of lines and dots is said to be less culturally loaded, or culture-reduced, as compared to say, a test based on vocabulary.  To put it crudely, people who speak English and those who speak Swahili are presumed to approach the first kind of test on an equal basis, but the second kind of test favors the speakers of whichever language the test is in.

2. The results above say that genetic variation explains more of the variation in the scores of culturally-loaded tests than it does of culturally-reduced tests.  The theory behind IQ actually requires it to be the other way around.   The paper argues (my crude rendition of the argument) that no, it can be explained as follows - across most environments,  genetically smart people master the culture-loaded stuff, while genetically less smart people don't, so culturally-loaded tests separate out genetically smart people and less smart people better than culturally-reduced tests.  In these very same set of environments, these very same sets of people, however, have the aspect of intelligence that cannot be learned (per IQ theory), as measured by culture-reduced tests, have less of the variation in test scores determined by their genetically smartness/less smartness.

3.  Since the above is not very clear, let me give you an analogy.  Let consider a general sports ability (say, agility), and the ability to play particular sports (say, tennis.).  The metaphor for IQ theory is that agility ought to be more heritable than prowess at tennis.   The finding is that, sorry, the statistics show it is the other way around.  The explanation is that people with  higher genetic ability in tennis tend to play tennis more and thus hone their tennis skills and the others don't, so that the variation in tennis ability correlates more strongly with genes;  while in the area of agility  there is less of this effect. It of course begs the quuestion of why avid tennis players are not correspondingly more agile.

(This metaphor fails because someone could be playing soccer and thus keeping agile. The theory of IQ does not admit of any such other avenue, however.)

4. But never fear, IQology will never fade or waver.  It is the Energizer Bunny of pseudo-sciences.

5. I can with fairly high confidence predict that I will be told that but for IQ studies we would not know that IQ theory is nonsense, and why am I against the march of knowledge?