Cortical dynamics of symbolic and non-symbolic numerical representations in young and older adults: an EEG study

Numeracy, the ability to understand and use numbers and basic numerical concepts, declines with age affecting everyday problem-solving, yet the underlying causes remain unclear. This study examined numerical cognition in young (24.1±4.73 years, n=31) and older adults (68.8±4.38 years, n=32), using a parity judgment task with three numerical formats: Arabic digits (N), Finger representations (F), and Dots (D). EEG (128 channels) was recorded. Accuracy was significantly lower for D than N and F (p = .02). Older adults responded slower than younger ones (p < .001), with both groups performing fastest for N (p < .001). Gender differences emerged for non-symbolic stimuli, where females responded slower than males (F, p = .04; D, p = .007), though no difference was found for N (p = .14). Multivariate pattern analysis revealed early (~200 ms) neural differentiation between N, F, and D, suggesting distinct processing for symbolic and non-symbolic representations. Older adults exhibited reduced neural decoding and increased latency. Source analysis showed lower activation and delayed responses in key numerical processing areas (intraparietal sulcus, fusiform gyrus) in older adults, alongside broader cortical recruitment. Among non-symbolic number representations, dots induce the lowest activation of the parietal cortex in the elderly. These findings suggest that age-related numeracy decline results from reduced efficiency in primary numerical processing areas, with compensatory activation of additional cortical regions. The study provides evidence for distinct neural networks supporting symbolic and non-symbolic number cognition and their age-related changes..