Neural and behavioral correlates of arithmetic development and learning in children

Abstract

Arithmetic learning improves mathematical competence, which is necessary for successful daily life. However, little is known about the neural underpinnings of arithmetic learning during childhood, the age when individuals learn most of the mathematical skills and the vast majority of our knowledge comes from adult studies. In this dissertation project, four studies were conducted to investigate the neural and behavioral correlates of arithmetic development and learning in children. In Study 1 arithmetic development was evaluated longitudinally to see whether it is monotonous or there are intermediate phases in which certain domain-general processes become important but disappear later. In Study 2 arithmetic complexity was evaluated to see whether it relies on both magnitude and cognitive processes, such as in adults. In Study 3 it was asked whether the findings in adults are valid for children or are there intermediate stages. Furthermore, it was evaluated whether few training sessions are reflective of more long-term learning processes. In Study 4 the brain activation changes during the course of learning were measured to see whether they reveal similar changes as in after arithmetic learning. The findings revealed that different domain-general cognitive processes are involved in different steps of arithmetic development and learning. Furthermore, arithmetic achievement occurs in two steps in children, first from slow effortful procedural processes to fast compacted procedural processes, and then to retrieval processes. These changes are distinguishable after one and several training sessions, and also during the course of learning. The findings are integrated in a theoretical model of arithmetic achievement in children, which contains two phases: (i) the efficiency increase (from slow effortful procedural processes to fast compacted procedural processes) and (ii) the strategy change (from fast compacted procedural processes to retrieval processes) phases. The model was developed based on two principles of brain function, optimum performance and energy consumption, and supported by several empirical studies. Taken together, this dissertation project provides a comprehensive framework for arithmetic development and learning in children. The findings might be helpful to develop educational and therapeutic interventions and also a new measure of intervention outcomes, particularly in individuals with mathematical learning disabilities.