A recent study indicates that genetic variants associated with higher educational attainment are also linked to lower levels of some behavioral and emotional difficulties in children and adolescents, particularly attention problems and stress-related physical complaints. These associations appear to run mainly through the part of the polygenic index tied to cognitive test performance rather than through its non-cognitive part, with the strongest patterns emerging in girls. The findings, published in Behavior Genetics, offer new perspective on how polygenic indices related to educational attainment track with youth mental health and everyday emotional development.
In behavioral genetics, researchers use a tool called a polygenic index, or PGI, to summarize statistical patterns across the human genome. Rather than measuring a fixed internal trait or an inborn biological essence, a PGI is a mathematical summary that adds up a vast number of tiny DNA associations identified in large population studies. When applied to complex outcomes like educational attainment, which reflects total years of formal schooling, these indices capture an intertwined mixture of genetic variation, social structures, and environmental opportunities.
Researchers often classify childhood behavioral challenges into two broad categories: internalizing and externalizing problems. Internalizing problems refer to difficulties directed inward, such as anxiety, sadness, social withdrawal, and physical complaints like stomachaches linked to emotional distress. Externalizing problems refer to outward-directed behaviors, such as rule-breaking and aggression, which disrupt home or classroom settings. Some areas, including attention problems, social problems, and unusual thoughts, fall between these two categories.
Untangling how genes relate to these behaviors within families is complicated because parents provide both their DNA and the home environment. In a 2018 study, researchers demonstrated that parental genetics influence children partly by shaping the rearing environment, a dynamic known as genetic nurture. Because of this environmental pathway, simply linking a child’s genes to their behavior can be misleading unless researchers account for parental DNA to isolate direct genetic associations.
At the same time, educational outcomes reflect many personal qualities beyond standardized test performance, such as persistence, curiosity, and emotional stability. Using a technique called GWAS-by-subtraction (GWAS stands for genome-wide association study, the large genetic surveys used to build these indices), a 2021 study split the genetic signal for educational attainment into two parts: one linked to performance on cognitive tests, and a “non-cognitive” remainder capturing everything else, such as traits tied to motivation or personality. This separation allowed scientists to test whether the genetics of educational attainment operate primarily through cognitive channels or through broader behavioral tendencies.
Separately, a 2020 study demonstrated that a child’s educational attainment PGI was related to having fewer attention and hyperactivity symptoms, even after controlling for parental DNA. Problem behavior is typically grouped with non-cognitive, or socioemotional, skills, which made it a natural test case for whether the decomposed PGIs relate to behavior in expected ways. This prompted the current research team to explore which specific genetic pathways associate with behavioral difficulties and whether these links differ by sex or informant.
“The educational attainment polygenic index (EA PGI) has become a very powerful predictor in sociogenomics, and is also being adopted into other fields like economics and sociology,” explained lead author Lyydia Alajääskö, a Ph.D. researcher in economics at Vrije Universiteit Amsterdam in the field of genoeconomics. “I was motivated to better understand what exactly the EA PGI signal contains.”
“From economic theory, we have two types of skills that predict later-life success, cognitive and non-cognitive skills,” Alajääskö told PsyPost. “I was able to test whether the cognitive/non-cognitive distinction holds up when you look at a behavior that should sit clearly on one side of it: childhood problem behavior. This is a novel test of this theory and a novel use of the decomposition approach.”
To investigate these questions, the research team analyzed genetic and behavioral records from the Lifelines study, a multi-generation health project based in the northern Netherlands. The analysis focused on 4,008 unique children and adolescents between 4 and 17 years old who contributed 7,787 total observations across multiple survey waves. All participants were of European genetic ancestry, and most observations relied on at least one parent’s genome being statistically reconstructed rather than directly measured.
Behavioral difficulties were evaluated using standardized surveys from the widely used Achenbach assessment system. Parents completed surveys evaluating their children’s behavior across 4,611 observations, while adolescents aged 13 to 17 completed self-report questionnaires across 3,176 observations. These surveys measured eight specific behavioral areas, ranging from anxiety and physical distress to attention difficulties and aggressive actions.
The authors constructed overall educational attainment PGIs for each participant and separated them into cognitive (Cog) and non-cognitive (NonCog) components. They estimated regression models that controlled for the child’s sex, age, birth year, survey wave, and broad ancestral background, applying a strict statistical threshold to guard against false discoveries. By comparing models with and without parental PGIs, the team separated the child’s direct genetic associations from indirect family-wide effects, calling an association robust if it held up after parents’ PGIs were taken into account.
The findings indicated that youth with higher overall educational attainment PGIs tended to show lower levels of some problem behaviors, but the robust associations were limited to a few areas. These associations ran mainly through the cognitive component of the PGI rather than the non-cognitive component. Specifically, higher values on the cognitive component were robustly linked to fewer attention problems in parent reports and fewer physical complaints tied to emotional distress in youth self-reports. The cognitive component was also associated with fewer anxious or depressed symptoms in parent reports, though this pattern was less consistent across model specifications.
By comparison, the non-cognitive component showed weaker and less consistent associations with youth behavior that were not statistically robust across the main models.
“However, this alone is not enough to conclude the non-cognitive pathway is insignificant or not relevant,” Alajääskö noted. “We could lack power to detect effects, and there are statistical limitations to the GWAS-by-subtraction decomposition in the first place. If anything, our results hint that the cognitive/non-cognitive distinction might partly reflect a shared underlying domain rather than fully separate pathways – though that remains speculative.”
Parents’ own PGIs were not statistically significant predictors of children’s problem behavior in the main models. This suggests the associations were not mainly explained by the home environments that parents’ genetics help shape, though the study had limited power to detect small indirect effects. The authors also note that even these direct associations are not purely biological, since they can include how children’s traits shape the responses they draw from others.
When the authors analyzed boys and girls separately, pronounced differences emerged between the sexes. Among girls, the cognitive component robustly predicted lower levels of anxious or depressed symptoms, unusual or obsessive thoughts, attention problems, and aggression in parent reports. In girls’ self-reports, the overall PGI was robustly linked to fewer physical complaints and internalizing symptoms, while the cognitive component held up only for physical complaints. For boys, several initial associations were present, but none remained statistically robust once parental PGIs were included in the model.
The type of survey respondent also mattered. In the full sample, the link between the cognitive component and fewer physical complaints was stronger in self-reports than in parent reports, while differences between informants were negligible for other behaviors. Among boys, no educational attainment PGI was significantly associated with self-reported behavior.
The authors suggested that cognitive skills and behavioral regulation might share an underlying neurocognitive foundation, such as executive functioning capacities like working memory and self-control. They emphasized, however, that because executive function was not directly measured, this explanation remains speculative.
Martin Fieder, an associate professor of evolutionary demography at the University of Vienna who was not involved in the research, described the work as methodologically sophisticated. “The main message is that the genetic association between educational attainment and childhood problem behavior seems to operate more through the cognitive component of the genetics of educational attainment (i.e., performance on cognitive tests) rather than through the ‘non-cognitive’ component (i.e., the remaining component of genetic variation associated with educational attainment).”
“Accordingly, genetic differences associated with better cognitive performance and higher educational attainment also appear to be associated with fewer behavioral problems, providing evidence for genetic overlap between cognitive performance and childhood behavioral regulation.”
“The study confirms previously reported genetic overlap between educational attainment and childhood problem behavior, but extends this work by showing that the association appears to be driven predominantly by the cognitive component of educational-attainment-related genetic variation,” Fieder observed.
The results are consistent with research covered by PsyPost in 2024, which found that higher intelligence was slightly associated with fewer behavioral and emotional difficulties, a link partly driven by shared genetic factors. While that earlier work evaluated directly tested IQ in seven-year-old twins, the current study examined genome-wide PGIs across children and adolescents aged 4 to 17.
As with all research, there are several caveats to keep in mind, particularly regarding how PGIs are interpreted. “This paper explicitly rejects notions of genetic determinism: our results don’t mean problem behavior is ’caused’ by genes, nor that outcomes are fixed,” Alajääskö emphasized. “PGIs are statistical summaries of genetic associations shaped by and reflecting the environmental and social structures in which genetic differences are expressed.”
“Relatedly, generalizability is a real question: genetic results in general are not representative of the world’s population, as there is a bias toward studies on individuals with European ancestry, partly driven by data availability,” Alajääskö added. “Secondly, genetic studies are also not representative of European country nationals: participants tend to be more educated, healthier, have higher socioeconomic status, and be slightly more female. For instance, our study uses a single cohort from a sub-region of the Netherlands, a country with low socioeconomic inequality and universal access to high-quality education.”
Beyond broad demographic questions, parental ratings introduce an additional layer of complexity, as ratings may partly reflect a parent’s own personality, stress levels, or expectations rather than objective child behavior. These parental perceptions can complicate comparisons across informants, especially when looking at internalizing symptoms like anxiety or physical complaints that children may feel more acutely than parents observe.
Evaluating such intricate genetic models also presents a distinct challenge for non-specialists. “A general reader is unlikely to be familiar with this kind of genetic analysis, and overall, the methods used are not easy to evaluate, even for someone like me who also works in behavioral genetics,” Fieder told PsyPost. “However, some of the authors are leading researchers in the field, which gives me confidence in the quality of the data analysis and the interpretation of the findings.”
Statistical power is another point to consider. While the sample was large enough to detect moderate direct genetic effects, it had more limited power to detect very small associations, which could partly explain why associations with the non-cognitive component were not statistically reliable.
Looking toward future research, Fieder emphasized the need for replication. “As always, I would like to see a replication study using an independent dataset,” he said.
The study, “Genetics of Educational Attainment and Childhood Problem Behavior: Direct Genetic Effects, Cognitive Pathways, and Sex Differences,” was authored by Lyydia Alajääskö, Marina Aguiar-Palma, Abdel Abdellaoui, and Titus Galama.
Leave a comment
You must be logged in to post a comment.