I somehow missed this thread at first. Reading through the comments here, I see several comments that express the correct idea, but without the key word, of "pl

I somehow missed this thread at first. Reading through the comments here, I see several comments that express the correct idea, but without the key word, of "pleiotropy," that is a gene (or gene assemblage) having more than one effect on phenotype. Some of the breeding examples mentioned in other comments already provide the warning: if we select human gene assemblages to maximize IQ scores (or something that can link even more directly to what we call "intelligence" in the real world outside the testing room), we may or may not produce healthy human beings, depending on what the pleiotropic effects of those gene assemblages are. Right now we just don't know. An article in the Nature Scitable collection explains what pleiotropy is in more detail.[1] I see that the author of the article kindly submitted here has just joined HN to join the discussion. I would respectfully suggest that the idea of "Crudely speaking, IQs of order 1,000, if the scale were to continue to have meaning," should be amended with a discussion of how IQ scores of 1,000 can't possibly exist, and surely wouldn't have meaning. That's because IQ scores are not derived from an interval scale, but rather an ordinal scale.[2] I read the story "Flowers for Algernon" when I was a kid too, and it was a very moving story, and the author did a reasonably good job of researching theories of psychology for his era. But his account of a mentally retarded (today we would say "intellectually disabled") adult having some kind of treatment that raised his IQ immensely is, at bottom, a made-up story, and the author was actually more clever than many of his readers in suggesting what is uncertain about what IQ scores mean. (He has some of his characters debate with one another about that, as I noticed when I reread the story in the last year.) We know Richard Feynman's IQ as a historical matter.[3] Feynman's IQ of 125 was above average, of course, but not astoundingly high. There are several Nobel Prize winners in physics with an IQ at a similar level, and plenty of physicists with higher IQs who never achieve as much in physics research as a Nobel laureate does. The uniform conclusion of ALL researchers who have looked at the relationship between IQ scores and scientific research achievement is that other personal qualities besides IQ matter a lot for research breakthroughs. One personal quality that is too little tapped by IQ testing is the habit of mind called "rationality." Amazing, that does not correlate especially strongly with IQ.[4] Improving "mindware," the ways that people think and how they reality-check their own thinking, may have a more powerful effect for human progress than merely increasing IQ. James R. Flynn, the dean of living researchers on human intelligence, suggests some strategies for improving thinking.[5] Being old enough to remember many gee-whiz promises about human genetic investigations that didn't pan out, for the moment I will bet on Flynn's approach providing more long-term benefit for humankind than the approach of looking for genome-wide associations with IQ.[6] [1] http://www.nature.com/scitable/topicpage/pleiotropy-one-gene-can-affect-multiple-traits-569 http://www.nature.com/scitable/topicpage/pleiotropy-one-gene... [2] See the references collected at https://en.wikipedia.org/wiki/Level_of_measurement#Ordinal_scale https://en.wikipedia.org/wiki/Level_of_measurement#Ordinal_s... or https://en.wikipedia.org/wiki/IQ_classification#Variance_in_individual_IQ_classification https://en.wikipedia.org/wiki/IQ_classification#Variance_in_... and note the diversity of cited authors, who all agree that IQ scores are ordinal rather than interval numbers. [3] https://en.wikipedia.org/wiki/IQ_classification#IQ_classification_and_genius https://en.wikipedia.org/wiki/IQ_classification#IQ_classific... [4] http://www.yalebooks.com/yupbooks/book.asp?isbn=9780300164626 http://www.yalebooks.com/yupbooks/book.asp?isbn=978030016462... [5] http://www.wiley.com/WileyCDA/PressRelease/pressReleaseId-104556.html http://www.wiley.com/WileyCDA/PressRelease/pressReleaseId-10... [6] "Most Reported Genetic Associations with General Intelligence Are Probably False Positives" http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3498585/ http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3498585/ "Understanding the Genetics of Intelligence: Can Height Help? Can Corn Oil?" http://apsychoserver.psych.arizona.edu/JJBAReprints/PSYC621/Johnson%20Current%20Directions%20Psych%20Science%202010%20(G%20and%20E%20in%20IQ).pdf http://apsychoserver.psych.arizona.edu/JJBAReprints/PSYC621/...