Developing, mature, and unique functions of the child’s brain in reading and mathematics

Abstract

Cognitive development research shows that children use basic “child-unique” strategies for reading and mathematics. This suggests that children’s neural processes will differ qualitatively from those of adults during this developmental period. The goals of the current study were to 1) establish whether a within-subjects neural dissociation between reading and mathematics exists in early childhood as it does in adulthood, and 2) use a novel, developmental intersubject correlation method to test for “child-unique”, developing, and adult-like patterns of neural activation within those networks. Across multiple tasks, children’s reading and mathematics activity converged in prefrontal cortex, but dissociated in temporal and parietal cortices, showing similarities to the adult pattern of dissociation. “Child-unique” patterns of neural activity were observed in multiple regions, including the anterior temporal lobe and inferior frontal gyri, and showed “child-unique” profiles of functional connectivity to prefrontal cortex. This provides a new demonstration that “children are not just little adults” – the developing brain is not only quantitatively different from adults, it is also qualitatively different.

Gender similarities in the brain during mathematics development

“…2) consistent with previous fMRI research on numerical cognition in children and adults. 15,16 Boys and girls showed equivalent mathematics-related neural responses (see Fig. 2).…”

“…a Percent signal change for number clips vs non-number clips in Study 1—redrawn data 15 (RIPS: t(25) = 4.01, p = 0.0005, 95% CI = 0.19–0.60, Cohen’s d = 0.79; LIPS: t(25) = 1.51, p = 0.26, 95% CI = −0.09–0.33, Cohen’s d = 0.23). b Preference for math clips vs non-math clips in Study 2—redrawn data 16 (RIPS: t(34) = 6.86, p < 0.0001, 95% CI = 0.12–0.23, Cohen’s d = 1.16; LIPS: t(34) = 4.12, p = 0.0002, 95% CI = 0.05–0.15, Cohen’s d = 0.70). c Slope of % signal change across increasing counting vs alphabet sequences in Study 3 (higher slope = greater sensitivity to sequence; RIPS: t(42) = 3.21, p = 0.003, 95% CI = 0.35–1.56, Cohen’s d = 0.49; LIPS: t(42) = 2.55, p = 0.014, 95% CI = 0.14–1.17, Cohen’s d = 0.39).…”

“…To compare the neural processes underlying mathematics development, we used functional magnetic resonance imaging (fMRI) to measure neural activity in 3–10-year-old children while they watched video clips that targeted early childhood mathematics skills (e.g., counting, addition; see Methods for more details). In total, 104 children (55 girls) participated in one of three natural viewing tasks (2 published studies 15,16 + 1 unpublished study, under review). Data were combined across natural viewing tasks by normalizing each subject to a within-task adult baseline using intersubject correlations.…”

“…The majority of analyses focus on similarities and differences in ‘neural maturity’, which indicates how well-developed and adult-like each child’s brain is during mathematical processing. 15,16 Neural maturity was calculated by conducting intersubject correlations 17 of the neural timecourses across the entire video between children and adults in every voxel of the brain (see Methods for details). Thus, it assesses the degree to which children’s neural activity resembled that of adults who watched the same video and allows the data to be combined across studies in a meaningful way.…”

Decision letter: A universal reading network and its modulation by writing system and reading ability in French and Chinese children

“…It is entirely conceivable that scanning children during the acquisition of reading would yield different results. The transient mechanisms of learning, in an immature and inexperienced brain, could be based on different mechanisms than those observed in the mature brain (Kersey et al, 2019). This is especially the case for a highly cultural, education-dependent activity such as reading (Dehaene and Cohen, 2007).…”