Chance Influences on IQ

Are there chance effects? Yes. How large are they? In most cases, we do not know. This page presents the evidence. In my research I have not seen a proper treatment of chance—perhaps because if it cannot be measured, there is nothing tangible to publish.

Non-biological randomness minimally affects IQ test results.

Perhaps one test taker catches a cold, meets a new girlfriend or boyfriend the night before test day, fights with a spouse, or breaks a valuable household item that morning. Such influences are undoubtably very low, not only because they would be rare among a large groups, and but also because they would affect the IQ test performance of the same person taking an IQ test twice.

Chance effects on IQ for identical twins:

How much of a person’s IQ is due to random events? That is, not due to DNA, to culture, or to environmental influences. To answer that question, one can start by looking at how much is not due to random events. The vast majority of identical twins who grew up together lived in comparable cultures and environments. In twin studies of the last century, adult identical twins raised together have correlations on their IQ test results of about 93% (a variance explained of 86%). (A twin will probably have a slightly higher result if they take the test again, but this affects both twins. Thus 14% of a raised-tiogether adult identical twin’s test results arises from chance or family environment. Perhaps one twin’s behavior, or the parent’s behavior affects the other’s ability. I’ll guess this at 4%. That leaves 10% for chance effects in the womb.

Percent of Hereditary IQ, Chance and Environment using Fraternal twins, and siblings:

The variance explained for identical twins raised apart is 0.69. The variance for identical twins does not include the variance of chance effects prior to the splitting of the egg. For siblings, this variance is included. There is, of course, the randomness in the selection of different chromosomes when the sperm and the egg are created, but without data it is impossible to estimate this. All we know is that the 0.25 is the subtraction of the random variance from the heredity variance.

The amount of influence on IQ test results due to heredity should be the same for everyone in a family. For argument’s sake, then, we can take the difference of 0.44 between identical twins and siblings and divide by 2, giving approximately .22 and argue that this is the chance component. We already estimated another chance component for identical twins of 0.10 (because of different IQ test results. This brings the total estimation of chance effects to 0.32. Evidence presented below suggests that this is a reasonable value.

If the above is correct, the amount of IQ test results due to heredity in a family (for one new generation) is 55%, the rest being due 32% to chance and 13% to environment and culture. Fifty-five percent is an interesting value, because it is only slightly higher than the amount of DNA that genomic studies have found, and mostly resolves the paradox of missing DNA.

While a few chance influence may permanently alter DNA, most chance influences, particularly womb influences, may not carry over to the next generation. They may be regarded as random, and be subject to regression.

Regression to the mean is a measure of biological randomness:

While there is rarely historical data on the IQ of women, there was, in England, good data on their heights. Sir Francis Galton, in the 19th century, collected data on hundreds of kinship relationships, including parents, grandparents and children, including twins. Galton discovered that stature (height) in each subsequent generation tended to regress toward the mean, the population average. He found that tall parents had, on average, less tall children, and that short parents had, on average, less short children. The same was true for tall or short siblings. And, although you may find it hard to believe, it was also true for the parents of tall or short children: The parents were more average. (Source 183).

Regression to the mean applies to IQ and all other human characteristics and all repeated measurements that are somewhat random. It is a statistical phenomenon, not a biological one. To understand this, consider that your extraordinary height, for example, is the result of a genetic component that is common in your family, plus a chance component. If the prevalence of the chance component follows a bell curve, in the next generation, or for your siblings or your parents, the chance component is likely to be more neutral.

If you and your spouse are smarter than most members of your families, it is likely that your children, on average, will be somewhat less smart than you, given equal nutrition and health. If you are dumber than most family members, your children are likely to be somewhat less dumb. If you are interested in the math and the of regression to the mean, see another explanation at Source 2.

The randomness component for height is about 33%, less in identical twins.

Galton found regression to the mean for height to be about 1/3. This very strongly implies that chance effects in height are about 33%. Is there a reason to believe that the causes of randomness for height are different from those for IQ? If the same influences apply, then, since we already have a random variation of .10 for identical twins’ IQ test results, about one-third of the .63 variance we have could be non-hereditary random, bringing the total to about .31.

Height is one of the most highly heritable human characteristics, and identical twins have very similar heights as they approach adulthood. (“Skeletal” in the chart in Source 3) For height, a study of approximately 80,000 identical twins, found variances explained due to chance (“non-shared environment”) of about 17% for boys and 24% for girls. (Source 4). Height is not affected by events such as bad events and sickness on test day, nor generally by different environments, so that differences in identical twin’s heights must be determined by biological chance events. Siblings, obviously, would have a larger chance component.

Other evidence for Biological Randomness that effects IQ

Evidence for large biological chance effects also comes from other mammals and other types of life. Armadillos (which are placental mammals that are born four at a time), have identical DNA, Yet the offspring have hormone levels that vary up to 10 times. I am taking the liberty of reproducing a chart from the study of Armadillos5. The armadillos were killed and analyzed immediately after birth.

Mice cloned for many, many generations — that is, mice that all have identical genes and whose parents and grandparents have the same identical genes — still differ in appearance, ability and emotional makeup.(Reference 6)

In the second half of the 20th century only two sources of random genetic variation were widely known: gene mutations and the shifting and copy number variation of DNA blocks during cell division. (We also now know that viruses and even our own RNA can rearrange sections of genes.) (Sources 7 8). Scientific work in the 21st century has focused on chance variations in DNA expression in fetuses and in later life. (Source 9 10 and their references) We now know that a large amount of randomness occurs both when DNA is copied, and when the copies (RNA) create proteins. (Sources 11 12  and their references.) (These chance differences can have important consequences for genetically influenced diseases, including cancer.) It seems likely that random modifications of DNA and RNA copying can account for 11% of the observed 11%variance in IQ found in all offspring.

Hormone variability can cause variations in IQ

We all know that hormones can have very powerful effects throughout the human body. Take, for example, the sex hormones; or consider growth hormone. Several hormones affect IQ,13 14 15 especially thyroid hormone (many articles on hypothyroidism).

Bacteria, viruses and protozoan may affect IQ

Our understanding of the role of bacteria and viruses in influencing our bodies is just emerging from the dark ages. There are over 3 million unique coding bacterial genes in the gut of the human body, compared to only about 20,000 in the human genome. (Source 16) Stomach bacteria, which differ from twin to twin (for both identical and fraternal twins), affect dietary preferences, mood and energy and even the progress of serious diseases. Many of these bacteria manufacture hormones and neurotransmitters that affect the brain. (Sources 17 18 19). Pathologists at the University of Alabama have even found, subject to proof of non-contamination, common stomach bacteria living in brain cells. (Source 20) and a brain microbiome is likely (Source 21)As for viruses and protozoa, consider that a parasite that infects mice can cause then to approach, rather than run from, cats. (Source 22) There are many other behavior-altering parasites. (Source 23)

Conclusion

In my website page on genome studies of heredity, I noted that practitioners claim to have found only one- one-half of all DNA responsible for IQ. That is because they have been looking for DNA responsible for the difference between identical twins and siblings, which used wrong data and faulty methodology. If one accepts that chance is responsible for much of the variance explained for IQ test results, then it follows that all or almost alll of the responsible DNA has been found. A paradox resolved!

Next page: https://iqscience.net/environmental-influences/

  1.  Book: Nicholas Wright Gillham, A Life of Sir Francis Galton, Oxford University Press, 2001. Note: All of the papers cited in this book are available on line, but I highly recommend this account of social science in the 19th century. ↩︎
  2. https://jaymans.wordpress.com/2015/10/21/regression-to-the-mean/ ↩︎
  3. https://pdfs.semanticscholar.org/aa70/4d096d7aff1cbf99c91764cc2c6ab0e5e4c2.pdf Polderman et al., Meta-analysis of the heritability of human traits based on fifty years of twin studies, nature genetics 2015 ↩︎
  4. https://www.nature.com/articles/srep28496  Genetic and environmental influences on height from infancy to early adulthood: An individual-based pooled analysis of 45 twin cohorts, jelenkovic, Sund et al, Nature, Scientific Reports ↩︎
  5. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC225138/pdf/pnas00121-0170.pdf A study of Monozygous Quadruplet Armadillos in Relation to Mammalian Interitance, Eleanor Storrs & Roger J. Williams, 1968 See also the references therein. ↩︎
  6. https://www.nature.com/articles/s41592-018-0224-7, Comparing phenotypic variation between inbred and outbred mice,Tuttle et al., Nature Methods, 2018. ↩︎
  7. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3821980/ RNA-mediated genome rearrangement: Hypotheses and evidence, Wenwen Fang1 and Laura F. Landweber2,*Bioessays, 2014 ↩︎
  8. https://www.ncbi.nlm.nih.gov/books/NBK9937/  The Cell: A Molecular Approach. 2nd edition ↩︎
  9.   https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2427204/ Phenotypically Concordant and Discordant Monozygotic Twins Display Different DNA Copy-Number-Variation Profiles Carl Bruder et al. , American Journal of Human Genetics 2008 ↩︎
  10. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3409253/ Gordon, Jihoon, et al, Neonatal DNA methylation profile in human twins is specified by a complex interplay between intrauterine environmental and genetic factors, subject to tissue-specific influence, Genome Research August 2012 ↩︎
  11. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3068611/, Balàzsi et al. , Cell, 2011 ↩︎
  12. https://royalsocietypublishing.org/doi/pdf/10.1098/rsif.2016.0833, Dattani & Barahona, Stochastic models of gene transcription with upstream drives: exact solution and sample path characterization ↩︎
  13. https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=2ahUKEwivp6XO_bviAhWMEBQKHVV4CvMQFjABegQIAxAC&url=https%3A%2F%2Fwww.neuroscience.cam.ac.uk%2Fpublications%2Fdownload.php%3Fid%3D20805&usg=AOvVaw2lkFuZEplxw3-MvkxYrhzh Prenatal and postnatal hormone effects on the human brain and cognition, Bonnie Auyeung et al., European Journal of Physiology, 2013 ↩︎
  14. https://academic.oup.com/cercor/article/19/2/464/346744 ,Sex Differences and the Impact of Steroid Hormones on the Developing Human Brain, Susanne Neufang, et al., Cerebral Cortex, 2019 * ↩︎
  15. https://www.ncbi.nlm.nih.gov/pubmed/17569675 Kupier et al., Heritability of reproductive hormones in adult male twins, Human Reproduction August 2012. ↩︎
  16. https://www.nature.com/news/2010/100303/full/news.2010.104.html, Gut bacteria gene complement dwarfs human genome, Bennett Hellman, 2010 ↩︎
  17. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4259177/ Leo Galland, The Gut Microbiome and the Brain, Journal of Medicinal Food, 2014. See many references for original studies. ↩︎
  18. https://www.researchgate.net/publication/23556365_A_core_gut_microbiome_in_obese_and_lean_twins, Turnbuaugh et al. Nature letters, January 2009 ↩︎
  19. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4915943/ Genetic determinants of the gut microiome in UK Twins, Goodrich et al, Cell Host Microbe, 2016 ↩︎
  20. https://www.sciencemag.org/news/2018/11/do-gut-bacteria-make-second-home-our-brains ↩︎
  21. https://pmc.ncbi.nlm.nih.gov/articles/PMC8165828/ ↩︎
  22. https://www.livescience.com/39772-parasite-makes-mice-unafraid.html ↩︎
  23. https://en.wikipedia.org/wiki/Behavior-altering_parasite ↩︎
Scroll to Top