An electrophysiological investigation of non-symbolic magnitude processing: numerical distance effects in children with and without mathematical learning disabilities.

INTRODUCTION: The aim of the present study was to probe electrophysiological effects of non-symbolic numerical processing in 20 children with mathematical learning disabilities (mean age = 99.2 months) compared to a group of 20 typically developing matched controls (mean age = 98.4 months). METHODS: EEG data were obtained while children were tested with a standard non-symbolic numerical comparison paradigm that allowed us to investigate the effects of numerical distance manipulations for different set sizes, i.e., the classical subitizing, counting and estimation ranges. Effects of numerical distance manipulations on event-related potential (ERP) amplitudes as well as activation patterns of underlying current sources were analyzed. RESULTS: In typically developing children, the amplitudes of a late parietal positive-going ERP component showed systematic numerical distance effects that did not depend on set size. For the group of children with mathematical learning disabilities, ERP distance effects were found only for stimuli within the subitizing range. Current source density analysis of distance-related group effects suggested that areas in right inferior parietal regions are involved in the generation of the parietal ERP amplitude differences. CONCLUSION: Our results suggest that right inferior parietal regions are recruited differentially by controls compared to children with mathematical learning disabilities in response to non-symbolic numerical magnitude processing tasks, but only for stimuli with set sizes that exceed the subitizing range.

Electrophysiological correlates of non-symbolic numerical magnitude processing in children: joining the dots.

Whether and in what way enumeration processes differ for small and large sets of objects is still a matter of debate. In order to shed light on this issue, EEG data were obtained from 60 normally developing elementary school children. Adopting a standard non-symbolic numerical comparison paradigm allowed us to manipulate numerical distance between stimulus arrays for different quantity ranges, i.e. the subitizing, counting and estimation ranges. In line with the existing literature, the amplitudes of parietal positive going ERP components showed systematic effects of numerical distance, which did not depend on set size. In contrast to the similarities in surface distribution of electrophysiological activity across all number ranges, applying source localization we found distance related current density effects in inferior parietal processing systems to be similar for all numerical ranges, there was, however, considerable variation in the involvement of medial parietal and lateral occipital regions. The precuneus, which is known to be involved in visual imagery, showed distance effects exclusively for numerical comparisons on large set sizes. In contrast, the processing of small quantities and stimulus arrays arranged into canonical patterns relied on lateral occipital areas that are linked to higher-level shape recognition. These findings suggest, on the one hand, that for explicit numerical decisions an involvement of domain-specific resources does not depend on quantity features of the visual input. On the other hand, it seems that the recruitment of mediating perceptual systems differs between the apprehension of small quantities and the enumeration of large sets of objects.