Cognitive ability becomes increasingly stable after age 3, study finds

An analysis of data from the Colorado Longitudinal Twin Study suggests that general cognitive ability tends to be highly stable across the lifespan, but this stability only emerges after about age 3. Measures of cognitive ability taken during infancy were associated with cognitive ability in adulthood, but only weakly. The study was published in PNAS.

Cognitive ability refers to a person’s capacity to acquire knowledge, think, reason, solve problems, and adapt to new situations. Individuals differ in these abilities, and such differences are meaningful even in early childhood, as they are linked to academic achievement. Higher cognitive ability is generally associated with better school outcomes. It also contributes to job performance, particularly in roles that require complex reasoning, quick learning, or problem-solving.

People with higher cognitive ability tend to make more informed decisions, as this capacity helps them weigh options and anticipate consequences more effectively. Cognitive ability also influences health-related behaviors, since individuals with higher ability are typically better at understanding medical guidance and managing long-term conditions. Research indicates that cognitive ability correlates with income, occupational status, and career advancement, making it a significant factor in socioeconomic mobility. It also shapes social interactions by supporting communication skills, perspective-taking, and conflict resolution.

Study author Daniel E. Gustavson and his colleagues aimed to examine the stability of general cognitive ability across the first 30 years of life. They assessed cognitive ability at five key time points: ages 1–2, 3, 7, 16, and 29. They hypothesized that cognitive ability would become increasingly stable after early childhood and that the predictive power of infant cognition for adult cognitive ability would be limited. The researchers also sought to quantify how early genetic and environmental influences on cognitive ability contribute to later cognitive outcomes by comparing data from twin participants.

The study analyzed data from 1,098 individuals in the ongoing Colorado Longitudinal Twin Study, including 552 females. While not every participant completed assessments at all five time points, the sample size remained substantial across waves: 640 completed infant cognition measures at 7 to 9 months of age as part of the Twin Infant Project; 814 completed one or more assessments between ages 1 and 2; 757 were assessed at age 3; 820 at age 7; 813 at age 16; and 684 participants completed the adult assessment at approximately age 29.

Free daily newsletter

As expected, the results showed that two specific infant cognition measures—novelty preference (object novelty) and tester-rated task orientation—were modestly associated with adult cognitive ability, with correlation coefficients of approximately 0.16 and 0.18. These associations were statistically significant but weak. In contrast, cognitive ability scores from age 7 onward showed strong correlations with later measures, with coefficients ranging from 0.72 to 0.86. This indicates a pattern of increasing stability with age.

Analyses using twin models revealed that 22% of the variation in adulthood cognitive ability could be attributed to genetic influences that were already present by age 3 or earlier. An additional 10% was explained by shared environmental influences measured at ages 1 to 2, which may include factors such as the home, school, or neighborhood environments shared by siblings.

“Findings suggest that genetic and environmental influences on GCA [general cognitive ability] demonstrate considerable stability as early as age 3 y [years], but that measures of infant cognition are less predictive of later cognitive ability,” the study authors concluded.

The study sheds light on the stability of cognitive abilities from infancy to adulthood. However, it should be noted that participants all came from Colorado. Results on other populations and cultures might not be identical. Authors also note that the reliability of infant cognition measures is much lower than that of cognitive ability measures used at older ages.

The paper, “Stability of general cognitive ability from infancy to adulthood: A combined twin and genomic investigation,” was authored by Daniel E. Gustavson, Giulia A. Borriello, Mohini A. Karhadkar, Soo Hyun Rhee, Robin P. Corley, Sally-Ann Rhea, Lisabeth F. DiLalla, Sally J. Wadsworth, Naomi P. Friedman, and Chandra A. Reynolds.