Cognitive Classes - What Is Intelligence?

The Intelligence Paradox: Why the Intelligent Choice Isn't Always the Smart One - Satoshi Kanazawa 2012

Cognitive Classes
What Is Intelligence?

Before we leave this chapter on intelligence, I'd like to introduce the concept of “cognitive classes,” which I use frequently throughout the rest of the book. It is a way of grouping individuals into five ordinal categories by their intelligence, from highest to lowest. This is a classification system that other scholars have invented25 and I have used before in my own work.26 The labels for the cognitive classes are used merely as a convenient shorthand, without any connotations.

Very bright (IQ > 125: roughly 5% of the US population)

Bright (110 < IQ < 125: roughly 20% of the US population)

Normal (90 < IQ 110: roughly 50% of the US population)

Dull (75 < IQ < 90: roughly 20% of the US population)

Very dull (75 < IQ: roughly 5% of the US population)

Here is a way to give you a quick flavor of what these cognitive classes mean. Among white Americans, 75% of those who earn a bachelor's degree are “very bright”; none are “very dull.” In contrast, 64% of those who drop out of high school are “very dull”; none are “very bright.”27


1. Gottfredson (1997a); Neisser et al. (1996)

[2]. Earlier attempts by intelligence researchers to dispel similar misconceptions about intelligence include Gottfredson (1997b, 2009) and Herrnstein and Murray (1994, pp. 1—24)

3. Burt et al. (1995); Profant and Dimsdale (1999)

4. Gottfredson (1997b); Jensen (1980)

5. Jensen (1980)

6. Jensen (1998, pp. 49—50)

[7]. Technically, heritability is the proportion of the variance in the trait across individuals that is influenced by genes. In order to have nonzero heritability, there has to be variance in the trait across individuals. So even though genes completely and entirely determine how many eyes you have, heritability of the trait “number of eyes” is 0, because there is no variance in the trait across individuals. All normally developing humans have the identical number of eyes

8. Turkheimer (2000)

[9]. Political attitudes: Alford, Funk and Hibbing (2005); Eaves and Eysenck (1974); divorce: Jockin, McGue and Lykken (1996); McGue and Lykken (1992)


11. Harris (1995, 1998); Rowe (1994)

12. Bouchard et al. (1990); Rowe (1994)

13. Deary et al. (2004)

14. Silventoinen et al. (2003)

15. Falconer (1960)

16. Cosmides (1989)

17. Daly and Wilson (1988, pp. 37—93; 1999)

18. Chomsky (1957)

19. Kanazawa (2004b)

20. Barash (1982, pp. 144—147)

21. Orians and Heerwagen (1992)

22. Miller and Kanazawa (2007, pp. 25—28)

[23]. If the problem was novel but recurring from then on (which thus ceases to be novel), then there would eventually be an evolved psychological mechanism specifically to deal with it. General intelligence would not be necessary to solve such problems. General intelligence is necessary only for novel and nonrecurrent adaptive problems

24. Gottfredson (1997a); Herrnstein and Murray (1994); Jensen (1998)

25. Herrnstein and Murray (1994)

26. Kanazawa (Forthcoming); Kanazawa and Perina (2009)

27. Jensen (1998, p. 296, Table 9.1)