Influence de la comparaison symbolique versus non symbolique sur le risque de difficultés d’apprentissage en mathématiques en éducation de la petite enfance

The main goal of this study was to clarify if students at risk of Mathematics Learning Disabilities had deficits in symbolic, non-symbolic processing, or both. Other purpose was to know if the Open Algorithm Based on Numbers (OBN) teaching method offers advantages over Closed Algorithms Based on Ciphers (CBC) method. A total of 307 Early Childhood Education students participated in this study. Based on student’ scores obtained in Test of Early Mathematics Ability (TEMA-3), that assesses formal and informal thinking, three analysis groups were designed: low-performance group, average-performance group, and high-performance group. There were significant differences in symbolic comparison and, consequently, the post hoc analysis showed significant differences between all groups in symbolic comparison except between the High and Average groups calculation. However, there were no differences in non-symbolic comparison between any of the groups. The teaching method and the mathematical competence showed a significant effect only on the symbolic comparison, resulting in a small size of effect for both the method and the Early Math Competence rate. These data support the dual representation hypothesis, in which the non-symbolic magnitudes would activate the Numerical Approximation System (ANS) and the symbolic quantities an exact representation of the numbers. This being a significant source of mathematical learning that could contribute to the improvement of the ANS.