Will Selecting Embryos for IQ Be the New Eugenics?

Two companies, Genomic Prediction and MyOme, claim they can screen embryos created using IVF for low intelligence (IQ) and cognitive disability.1 These companies do this by making use of large datasets and sophisticated algorithms to identify a pattern of genetic markers that correlate to cognitive abilities. In a statement, Genomic Prediction co-founder Stephen Hsu said the company will not use this technology to select for high IQ, but only against mental disabilities. But he points out that eventually someone will use this technology to select for intelligence.2

Genomic Prediction and MyOme use something called a "polygenic risk score" to identify an embryo's risk for a cognitive disability. Polygenetic risk scores are used by companies such as 23andMe, that do commercial genetic testing for adults.3 Genomic Prediction and MyOme use preimplantation genetic diagnosis to apply this technology to embryos. Genomics Prediction admits that its methods are not able to accurately predict each embryo's IQ, but it can pinpoint genetic outliers among embryos that might indicate a low IQ. Parents can then choose whether to implant such an embryo or discard it.

Of course, because all of a given couple's embryos are made from the same genetic material, parents may be disappointed with all of their available choices. For instance, a particular couple's genetic material might produce four embryos, all of which have a propensity for Type I diabetes or a sub-average IQ. One can see, then, how a market for premium eggs and sperm would exist among Genomic Prediction's and MyOme's clientele. Couldn't their large data sets be used to create a searchable database of donors with desirable traits?


The fundamental assumption behind Genomic Prediction's and MyOme's technique is that complex traits, including disease susceptibility, mental capacity, and personality, are largely determined by one's genes. This is the premise of a field of study called sociogenomics. Sociogenomics, also called sociobiology, began as a study of the molecular basis for insect social behavior.4 Once genetic sequencing became cheap and easy, scientists could map the entire genome of an organism. This allowed researchers to collect enough data to correlate social behavior with patterns of genetic variance within the genome. The genetic variances are called single-nucleotide polymorphisms (SNPs), which gives you an idea of just how small these variances are—the difference of a single base letter in DNA. Today, algorithms sift through extensive amounts of data to correlate certain behaviors or traits with the patterns of hundreds or thousands of SNPs in a process descriptively called "genome-wide association studies" (GWAS).

Same Technology, Opposite Approaches

Analyzing large swaths of genetic data has helped many individuals born with rare genetic diseases. These diseases may be caused by a single mutation within the entire genome. The best way to find the mutation is through efficient genome sequencing combined with the use of sophisticated algorithms to sift through large data sets. Once the mutation is identified, researchers can try to develop a gene therapy that silences it.

When hunting for the genetic culprit of a rare disease, researchers start with a hypothesis—these symptoms are caused by a genetic mutation—and then proceed to find where those mutations occur by comparing the patient's genome to thousands of other genomes. If the researchers find more than one patient with the same symptoms and the same mutation, then their hypothesis is tentatively confirmed. If they find several patients with the same symptoms and the same mutation, they are more confident of their hypothesis. If they do not find a match, then their hypothesis is not confirmed.

Sociogenomics employs the same technology as rare disease research, but it uses a different philosophical method. Nathaniel Comfort, a historian of biology and medicine, says the problem with sociogenomic studies is they don't start with a hypothesis; they start with the data.5 Sociogenomics doesn't look for single mutations, but for patterns of genetic variations, which can include mutations, and it correlates those patterns to particular traits. These traits are complex and are likely caused by both genetic and environmental factors.

Rather than making predictions at the beginning of the scientific process, sociogenomics researchers make predictions at the end, after an algorithm matches patterns of SNPs to complex traits. However, Comfort says that with such large data sets, "the number of possible combinations of SNPs is so large that finding associations with any given trait is practically inevitable." In other words, if you look hard enough, you'll find a pattern of SNPs to correlate with anything.

These are two different methods of reasoning that use the same techniques. The search for whether a disease is caused by a mutation is closer to the traditional scientific method, which uses deductive reasoning. The hypothesis can be confirmed or not. Sociogenomics, on the other hand, is inductive reasoning. It looks at patterns and draws generalized conclusions, creating the potential for false positives.

Problems with Measuring Intelligence

The idea that intelligence can be measured, let alone selected for, poses another problem with screening embryos for intelligence. French psychologist Alfred Binet invented the first iteration of the IQ test in the late nineteenth century to help teachers discern which students needed additional help in the classroom. Binet assumed that a person's intelligence developed with age, but that his intellectual level relative to that of his peers stayed the same. In 1905 Binet developed the Binet-Simon scale, which became the basis for the Stanford-Binet scale used to measure IQ in the United States.6 While Binet intended his test as an educational aid, it came to be used for more nefarious purposes. Eugenicists used it as a tool for identifying and sterilizing "feebleminded" people. The test was also used to justify racial prejudices: African-Americans and certain non-English-speaking immigrants were deemed less intelligent based on their scores on the IQ test—a test written in English that included culturally specific questions.

Critics at the time questioned whether intelligence is really static throughout a person's life, a key assumption for eugenicists and an implied assumption in today's embryo screening. Henry Goddard, a eugenicist who brought the Binet test to America from France said, "The degree of brightness of an individual is expected to remain approximately constant. Were this not so, prognostication would be impossible. Intelligence testing would be of little value."7

In fact, a study conducted by Harold Skeel and Harold Dye in the 1930s demonstrated that IQ among children was variable and that IQ was more an indication of a person's particular situation than an innate quality. Their study began with two baby girls who were institutionalized because their IQ scores were in the 40s (100 is average). However, when the babies were transferred to a women's institution and provided with care, attention, toys, and affection, their scores increased by at least 35 points. After several more months, their IQ scores were near average.

Skeel and Dye conducted another study, this time with thirteen children who were transferred from an orphanage to a more caring environment. They were compared to a control group that stayed at the orphanage. All thirteen children in the nurturing environment improved their IQ scores from the "feebleminded" level to near average. The control group did not. Unfortunately, this study was ignored and dismissed as "bad science."8

Biological Determinism & Reductionism

Beyond the problematic assumption that IQ stays the same throughout a person's lifetime is the assumption that intelligence is even measurable at all. IQ hinges on the notion that intelligence is quantifiable; a notion that Stephen Jay Gould questioned in his book The Mismeasure of a Man.

In the late nineteenth and early twentieth centuries, there was a push to quantify everything about human beings, including their mental abilities. This endeavor was known as psychometrics. Gould, though himself a proponent of Darwinian evolution, was an outspoken critic of psychometrics and the fields that employed it: sociobiology and evolutionary psychology, both of which rely on Darwin's theory of natural selection. Gould criticized biological determinism as scientifically spurious and politically motivated.9

In reality, intelligence is not reducible to a single measurable trait. Alfred Binet himself thought his test did not give the full picture of intelligence because it didn't provide a good measure of things like creativity or "emotional intelligence."10 Proponents of multiple intelligences have shown that IQ tests tend to be culturally specific and neglect other types of intelligence that are advantageous in different contexts.

We see evidence against biological determinism in cases where a gene is known to increase one's chances of contracting a particular disease—as the BRCA gene does for breast cancer, for instance—yet a person known to have the gene never gets the disease. Various other factors besides genetics, such as one's environment and lifestyle, determine whether or not a gene will be expressed. Thus, while genetics can inform probabilities, it cannot provide certainty.

Genome-wide association studies are even more tentative than gene-to-disease studies. GWAS can show correlations, but there is no way of knowing if a pattern of SNPs causes a particular trait, such as cognitive disability, or not. Nathaniel Comfort says it is like noticing that people with Down syndrome tend to be shorter than average. But just because someone is short, it does not mean that he has Down syndrome. Similarly, there is no way to tell whether a certain pattern of SNPs observed in an embryo is actually related to cognitive ability or not.

Embryo screening is, therefore, simply a form of modern-day eugenics dressed up in the trappings of technology and "parents' choice" lingo. Embryos that do not pass genetic muster are discarded because it is thought that they won't measure up to cultural notions concerning the value of intelligence and other desirable traits for making life worthwhile. Embryo screening is not therapy. It is a justification for getting rid of a potential "problem" by getting rid of the person.

1. Bob Yirka, "Progress in genetic testing of embryos stokes fears of designer babies," Medical Xpress (Nov. 16, 2018): https://medicalxpress.com/news/2018-11-genetic-embryos-stokes-babies.html.
2. Clare Wilson, "Exclusive: A new test can predict IVF embryos' risk of having a low IQ," New Scientist (Nov. 15, 2018): newscientist.com/article/mg24032041-900-exclusive-a-new-test-can-predict-ivf-embryos-risk-of-having-a-low-iq.
3. Antonio Regalado, "Eugenics 2.0: We're at the Dawn of Choosing Embryos by Health, Height, and More," MIT Technology Review (Nov. 1, 2017): technologyreview.com/s/609204/eugenics-20-were-at-the-dawn-of-choosing-embryos-by-health-height-and-more.
4. Gene E. Robinson et al., "Sociogenomics: Social life in molecular terms," Nature Reviews Genetics (April 2005): researchgate.net/publication/7973898_Sociogenomics_Social_life_in_molecular_terms.
5. Nathaniel Comfort, "Sociogenomics is opening a new door to eugenics," MIT Technology Review (Oct. 23, 2018): technologyreview.com/s/612275/sociogenomics-is-opening-a-new-door-to-eugenics.
6. Con Stough, "Show us your smarts: a very brief history of intelligence testing," The Conversation (Oct. 9, 2015): https://theconversation.com/show-us-your-smarts-a-very-brief-history-of-intelligence-testing-45444.
7. Quoted in Anya Kamenetz, The Test: Why our schools are obsessed with standardized testing—but you don't have to be (Public Affairs, 2015), 55.
8. Ibid., 56–57.9. See the Introduction to the 2nd edition of The Mismeasure of Man, by Stephen Jay Gould (W. W. Norton & Co., 1996).10. Daphne Martschenko, "The IQ test has a dark history" The Week (Oct. 24, 2017): https://theweek.com/articles/731957/iq-test-dark-history.

has an M.S. in chemistry from the University of Texas at Dallas, and an M.A. in bioethics from Trinity International University. She resides in Dallas and currently works as a freelance science writer and educator.

This article originally appeared in Salvo, Issue #48, Spring 2019 Copyright © 2020 Salvo | www.salvomag.com https://salvomag.com/article/salvo48/gene-shopping