The two-network framework of number processing: a step towards a better understanding of the neural origins of developmental dyscalculia

Abstract

Developmental dyscalculia is a specific learning disorder that persists over lifetime and can have an enormous impact on personal, health-related, and professional aspects of life. Despite its central importance, the origin both at the cognitive and neural level is not yet well understood. Several classification schemas of dyscalculia have been proposed, sometimes together with an associated deficit at the neural level. However, these explanations are (a) not providing an exhaustive framework that is at levels with the observed complexity of developmental dyscalculia at the behavioral level and (b) are largely mono-causal approaches focusing on gray matter deficits. We suggest that number processing is instead the result of context-dependent interaction of two anatomically largely separate, distributed but overlapping networks that function/cooperate in a closely integrated fashion. The proposed two-network framework (TNF) is the result of a series of studies in adults on the neural correlates underlying magnitude processing and arithmetic fact retrieval, which comprised neurofunctional imaging of various numerical tasks, the application of probabilistic fiber tracking to obtain well-defined connections, and the validation and modification of these results using disconnectome mapping in acute stroke patients. Emerged from data in adults, it represents the endpoint of the acquisition and use of mathematical competencies in adults. Yet, we argue that its main characteristics should already emerge earlier during development. Based on this TNF, we develop a classification schema of phenomenological subtypes and their underlying neural origin that we evaluate against existing propositions and the available empirical data.

Changes in the superior longitudinal fasciculus and anterior thalamic radiation in the left brain are associated with developmental dyscalculia

“…Additionally, there have been studies suggesting different brain circuits for verbal and visual number processing, i.e., magnitude comparison and arithmetic problem solving tasks including exact (fact retrieval) and approximate calculations (Dehaene, 1992;Dehaene and Cohen, 1995;Amalric and Dehaene, 2019). Similarly, possible subtypes of DD can be reflected by some deficits/alterations in common and/or dissociable numerical networks in the brain as previously suggested (Stanescu-Cosson et al, 2000;Holloway et al, 2010;Lyons et al, 2015;Klein and Knops, 2023). We had no focus on any possible DD subtypes in our study.…”