Left angular gyrus disconnection impairs multiplication fact retrieval

Arithmetic fact retrieval has been suggested to recruit a left-lateralized network comprising perisylvian language areas, parietal areas such as the angular gyrus (AG), and subcortical structures such as the hippocampus. However, the underlying white matter connectivity of these areas has not been evaluated systematically so far. Using simple multiplication problems, we evaluated how disconnections in parietal brain areas affected arithmetic fact retrieval following stroke. We derived disconnectivity measures by jointly considering data from n=73 patients with acute unilateral lesions in either hemisphere and a white-matter tractography atlas (HCP-842) using the Lesion Quantification Toolbox (LQT). Whole-brain voxel-based analysis indicated a left-hemispheric cluster of white matter fibers connecting the AG and superior temporal areas to be associated with a fact retrieval deficit. Subsequent analyses of direct grey-to-grey matter disconnections revealed that disconnections of additional left-hemispheric areas (e.g., between the superior temporal gyrus and parietal areas) were significantly associated with the observed fact retrieval deficit. Results imply that disconnections of parietal areas (i.e., the AG) with language-related areas (i.e., superior and middle temporal gyri) seem specifically detrimental to arithmetic fact retrieval. This suggests that arithmetic fact retrieval recruits a widespread left-hemispheric network and emphasizes the relevance of white matter connectivity for number processing.

Arithmetic Fact Retrieval

When diagnosing children with learning disorders (as per ICD-10), their scholastic performance has to be significantly below the level of intelligence. Although this discrepancy criterion has received much criticism in the field of literacy, few researchers in mathematics have examined it. We used a two (mathematical performance) by two (intelligence) factorial design to analyze the arithmetic fact retrieval of low-performing children in mathematics who met the criterion (developmental dyscalculia) or did not (mathematical difficulties) and of two groups of average-achieving children matched for intelligence. The four groups (each n = 27 third-graders) were matched for their attention span and their literacy skills. Children solved addition verification tasks with numbers up to 10 and 20 under standard and under dual task conditions requiring further working memory capacity to evaluate the potential use of counting strategies. Performance in addition tasks proved to be associated with mathematical achievement especially in the higher number range, whereas dual task performance did not point to the use of counting strategies among low performers in mathematics. No interaction between mathematics and intelligence was identified, which would have confirmed the discrepancy criterion. These results illustrate that stable knowledge of arithmetic facts is essential for mathematical achievement, regardless of whether the discrepancy criterion is met.