It has been difficult to determine how cognitive systems change over the grand time scale of an entire life, as few cognitive systems are well enough understood; observable in infants, adolescents, and adults; and simple enough to measure to empower comparisons across vastly different ages. Here we address this challenge with data from more than 10,000 participants ranging from 11 to 85 years of age and investigate the precision of basic numerical intuitions and their relation to students' performance in school mathematics across the lifespan. We all share a foundational number sense that has been observed in adults, infants, and nonhuman animals, and that, in humans, is generated by neurons in the intraparietal sulcus. Individual differences in the precision of this evolutionarily ancient number sense may impact school mathematics performance in children; however, we know little of its role beyond childhood. Here we find that population trends suggest that the precision of one's number sense improves throughout the schoolage years, peaking quite late at ∼30 y. Despite this gradual developmental improvement, we find very large individual differences in number sense precision among people of the same age, and these differences relate to school mathematical performance throughout adolescence and the adult years. The large individual differences and prolonged development of number sense, paired with its consistent and specific link to mathematics ability across the age span, hold promise for the impact of educational interventions that target the number sense.aging | analog magnitude | approximate number system | cognitive development | ensemble representation A lthough the particulars of our minds may differ from person to person, some aspects of cognition are close to our corethey are universally shared, present in the young, and actively engaged throughout our lifetimes (1, 2). Investigating developmental changes in these core systems may present us with a picture of how the mind transforms from infancy to senescence. Here we investigated change in the approximate number system (ANS), the cognitive system that gives rise to our basic numerical intuitions (3). The ANS generates nonverbal representations of numerosity in nonhuman animals (4, 5), infants (6, 7), school-aged children (8-10), and adults from mathematically fluent cultures (11,12) as well as cultures that do not practice explicit mathematics (13,14). In humans, imaging results suggest that these basic intuitions are supported by neurons in the intraparietal sulcus (15-18), a role that can be observed shortly after birth (19). Given the phylogenetically widespread occurrence of this primitive cognitive resource, the ANS might make little or no contact with the formal mathematical abilities that humans struggle to master and that no other animals acquire (20). Alternatively, this system may be a critical foundation upon which formal mathematical abilities are constructed (21,22). Although some evidence suggests a link between the ANS and formal mathemat...