Infants are thought to possess an innate specific capacity to process numerical information. In this article, we review the past research that has focused on unveiling the timing and localization of the related brain mechanisms with the purpose of depicting a neurodevelopmental blueprint of this capacity from birth. A systematic search of studies published between 1998 and 2023 was conducted. A total of 21 studies with 732 participants (age rage: 30 weeks of gestation to 6 years) met the study selection criterion. EEG, fMRI and fNIRS studies consistently support the existence of brain responses (mainly in the right parietal, bilateral frontal and occipital cortex) that reflect sensitivity to numerical features even before birth. These enable the infant brain to code numerical information independently of other non-numerical magnitude dimensions. Small (<4) or large (>4) numerosities seem to diverge in dissociable brain responses from the second semester of life, suggesting a neurodevelopmental specialization. Variations in the brain’s sensitivity to numerical information across participants and whether they can anticipate the individual’s development of future numerical skills remains uncertain, due to the scarcity of longitudinal studies. Understanding how familial and other contextual factors shape these initial biological predispositions and give rise to typical and atypical trajectories requires further investigation.
Everyone agrees that environments influence learning, but only a small percentage of studies of cognitive development relate children’s specific learning environments to their learning. In this article, I examine relations to math learning of two types of specific learning environments: math textbooks and home math environments. The strength of the relations appears to differ for the two types of environments; characteristics of math textbooks seem to play an important role in shaping school-age children’s arithmetic with whole numbers, fractions, decimals, and percentages, but characteristics of home math environments seem to be only weakly related to preschoolers’ math knowledge. The differences cannot be attributed entirely to differences in the children’s ages or to difficulties in measuring preschoolers’ mathematics knowledge; studies relating preschoolers’ literacy environments to their reading comprehension yield much stronger relations, and measurement of preschoolers’ math knowledge is sufficiently valid to yield substantial relations to the same children’s math knowledge in elementary and high school. Two types of issues are identified that seem to contribute to the differing relations of the specific learning environments to children’s mathematics knowledge: issues involving measurement of specific learning environments and influences of unmeasured variables. Relating learning environments to learning is likely to be useful for understanding cognitive development in other domains as well.