z53\doc\web\2001\10\birthead.txt (I don't endorse it, but it's interesting...) Philippe Rushton [rushton@uwo.ca] 10/2001 writes I hope the following is useful for David Cohen's question about small head sized babies and later IQ scores: The overall correlation between head circumference at birth and IQ at age 4 (and again at age 7) is 0.20. There is a lot of individual variance in the strength of that relationship. In my own last analysis of these kind of data (in the 1997 issue of Intelligence, copies sent on request) there were two children (both boys) who were born with birth weights and head circumferences more than 3 standard deviations below the mean and were transferred to wards for the premature. The two outliers provide a useful illustration of the phenomenon of "catch-up growth." At birth, the two boys averaged 1.816 kilograms in body weight with a head circumference of 29.25 cm -- both measures 3 standard deviations below the mean. However, these disadvantages (for example, in cranial size) were reduced to -0.77 SD by 4 months, to -0.60 SD by 1 year, and to only -0.10 SD by 7 years. At age 7, one boy had an IQ of 115 and the other of 111. The phenomenon of "catch-up growth" following deficits caused by malnutrition or illness demonstrates that development is guided by constant self-corrections until some targeted end-state is reached. Deprived children, if moved to a satisfactory environment, subsequently develop very rapidly. They regain the growth trajectory they would have been on if the diversion had not occurred, following which growth slows down and development proceeds at the normal pace. Developmental processes are constantly involved in a match-to-model process with an inherent growth equation. From the Discussion section of that article..... The data on catch-up growth, when applied to the Asian--White--Black mean differences in both brain size and IQ, supports the genetic hypothesis of the race differences.argues in favor of the existence of genetic factors. The reason for this is that by age 7, Asians are above average in brain size but below average in body size whereas the opposite pattern is found for Blacks and an intermediate pattern is found for Whites. If major environmental insult or nutritional deprivation was the cause of the differences, as has often been posited for low birth-weight infants, one would not expect the particular pattern of catch-up growth that is demonstrated by these data. Moreover, malnutrition does not usually influence brain weight, suggesting that evolution has selected brains to be especially well conserved. Additional recent evidence for a genetic contribution to racial differences in mean IQ can be found in Jensen (1998; in press), Levin (1997), Lynn (1997), Rowe and Cleveland (1996), and Rushton (1997). One of the major findings of the Collaborative Project has been that low birth-weight babies are not as much at risk as has often been supposed. Broman (1989) reviewed the overall findings pointing out that although mean raw scores at 8 months of age on the Bayley Scales increased linearly with birthweight as predicted, with the lightest group (<2,000 grams) scoring about two standard deviations below the heaviest group (>3,501 grams), birthweight explained only 5% to 6% of the variance in test scores in this population of 31,000 infants. By age 4, in a regression analysis with six covariates (gestational age, birthweight, head circumference, sex of child, ethnicity, and maternal education), birthweight explained less than 1% of the variance in IQ scores. Ethnicity and maternal education were the best predictors, accounting for 16% and 6% of the variance, respectively. In contrast to birthweight, head circumference at birth was frequently retained in multivariate analyses, indicating a greater independent contribution to cognitive outcome at both ages (Broman, 1989). These results show that birthweight has less of an effect on measures of cognitive development at age 4 years than at 8 months. The most important predictors of cognitive performance were characteristics of the family.