As genetic data expand, researchers urge caution in how predictors of education outcomes are used

It has long been known that the number of years a person spends in school is heritable. More recent research focuses on the links between variation in specific genetic markers and educational attainment. After identification of such markers, follow-up work has examined associations between differences in these genetic markers and differences in the development of traits such as cognitive ability, school performance, verbal ability, and learning disabilities. A 2018 study found that a composite summary of an individual's genetic makeup, a "polygenic score," explains roughly 12 percent of variation in years of schooling and 9 percent of variation in cognitive ability.

In a review published online today in AERA Open, a peer-reviewed journal of the American Educational Research Association, researchers from Stanford University and the University of Cambridge warn that -- as the predictive power of genes tied to learning and educational outcomes increases and access to genetic data expands -- researchers, educators, and policymakers must be cautious in how they use such data, interpret related findings, and, in the not-too-distant future, apply genetics-informed student interventions.

The review, titled "Genetics and Education: Recent Developments in the Context of an Ugly History and an Uncertain Future," was authored by Daphne Martschenko, a PhD candidate at the University of Cambridge; Sam Trejo, a doctoral candidate at Stanford University; and Benjamin W. Domingue, an assistant professor of education at Stanford University.

"The field of molecular genetics is moving fast and the floodgates of genetic data have opened," said Domingue. "It is important that researchers, policymakers, practitioners, and the public be aware of the limitations of genetic data and the corresponding dangers of misuse."

"In the past, genetic data were used to deny children opportunities for educational growth, erode public support for social policies, or propagate racist or classist beliefs," Domingue said. "Genetic data are a technological innovation that may allow for improvements in the human condition, but it is critical that we ensure such innovations don't come at the expense of affording all children the opportunity to learn."

The authors note that genetic research, as part of an "ugly history" of bigoted ideologies, was used to bolster unjust policies in the United States. The use of genetic language to describe racial and socioeconomic differences in cognitive ability and academic performance was commonplace in the 19th and 20th centuries and continues today, according to the research team.

The authors also note that:

• Research on human genetics might offer information about developmental trajectories of traits -- such as ADHD, dyslexia, and other learning disabilities -- that are relevant to education. In the near term, such information may offer insight into strategies for intervention. In the longer term, it may even be useful for individual-level risk prediction. However, there is no evidence to suggest that any child cannot thrive in the appropriate environmental setting -- and there is no research that justifies depriving children of needed interventions.
• Genetic inheritance is not destiny. An increased understanding of how one's genetic inheritance influences development does not change the fact that environments play a massive role in educational attainment and success. Genes are implicated in many developmental processes, but environments have large effects on all people independent of their genetic makeup and may even moderate the role of genetic influences in some cases. Helping those students most at risk for health or sociobehavioral problems is one of the most pressing demands in education, irrespective of why such students are at risk. Providing an environment rich with educational possibilities for all children should always be a top priority.
• Genetics may provide valuable insight into educational processes, but it cannot invalidate the importance of well-crafted public policy needed to support educators determined to address the collective task of educating the next generation.

While the predictive power of genetic data is still too weak to be used for individual-level interventions, such approaches -- involving, perhaps, specialized interventions for a child with a genetic profile that suggests possible difficulties with reading -- will likely be in demand someday soon, according to the authors.

"At some point in the future, parents of a young student may show up in a principal's office, armed with genetic data indicating that their child is at a heightened risk for dyslexia, and ask that learning supports be put into place immediately," said Domingue. "In envisioning this scenario, many questions come up. Do such learning supports exist? Do we know them to be effective? And might they be useful for all students?"

As genetic data become more commercialized, with higher income people having greater access to the information, educators and policymakers will have to consider the implications of using genetic predictors for both efficacy and equity, according to the researchers.

Equity concerns are especially relevant since genetic prediction can vary by ancestry. Research has found that the predictive power for European ancestry-derived polygenic scores can be lower in people of non-European ancestry. To date, research on genetic effects has been conducted largely on people of European ancestry.

"Will all children be able to take advantage of any useful diagnostic information that becomes available?" asked Domingue. "We would argue that to ensure that genetics research benefits all and not just some, regulation meant to safeguard privacy and equal treatment will be increasingly important."