Numerical comparison is a primary measure of the acuity of children's approximate number system (ANS). ANS acuity is associated with key developmental outcomes such as symbolic number skill, standardized test scores and even employment outcomes (Halberda, Mazzocco, & Feigenson, 2008;Parsons & Bynner, 1997a). We examine the relation between children's performance on the numerical comparison task and the number line estimation task. It is important to characterize the relation between tasks in order to develop mathematics interventions that lead to transfer across tasks. We find that number line performance is significantly predicted by non--symbolic comparison performance for participants ranging in age from 5 to 8 years. We also evaluate, using a computational model, if the relation between the two tasks can be adequately explained based on known neural correlates of number perception. Data from humans and non--human primates characterizes neural activity corresponding to the perception of numerosities. Results of behavioral experimentation and computational modeling suggest that though neural coding of number predicts a correlation in participants' performance on the two tasks, it cannot account for all of the variability in the human data. This is interpreted as consistent with accounts of number line estimation in which number line estimation does not rely solely on participants' numerical perception. Introduction Children in school face the daunting task of learning a complex number system. Children must learn to manipulate numbers in increasingly sophisticated ways to stay on track in the formal education system. Understanding how different aspects of numerical cognition are related to one another within learners is an important goal for improving educational outcomes. Numerical cognition research has focused on performance on a few model tasks, such as number comparison and number line estimation. Numerical comparison is used to measure number sense, learners' fluency with numbers. The number line estimation task measures number to space mapping ability, which has been shown to be important in numerical development (Gunderson, Ramirez, Beilock, & Levine, 2012). It is unclear the degree to which performance on these two tasks relies on the same mechanisms. A central question is whether individual differences in performance across these tasks is primarily due to a shared cognitive mechanism.
Study MotivationThe current study addresses the questions, (1) does children's non--symbolic comparison skill predict performance on the symbolic number line task when accounting for age, (2) if so can a computational model of the neural coding of number account for the relation. Several cognitive representations have been used to characterize participants' behavior, such as the mental number