How do children learn number words? Comparing proposed mechanisms for number word acquisition

• In Event: Talking About Numbers

Abstract

Math abilities are related to lifelong career, health, and financial outcomes (Agarwal & Mazumder, 2013; Currie & Thomas, 2001; Reyna & Brainerd, 2007). Individual differences in math performance are present early in life and are remarkably stable throughout development (Duncan et al., 2007). Understanding number words, a foundational math skill, is critical for the development of later symbolic math abilities (Geary et al., 2018). Number word knowledge involves mapping the word label for each number to a set of items of the specified size. Mature number knowledge is defined by understanding the cardinality principle—that each number word refers only to an exact set of that quantity with the last number in the count list referring to the total number of items in the set (Carey, 2009; Wynn, 1990). Extensive discussion has centered around how children learn to map number words to their referred quantities (see Sella et al., 2021; Carey & Barner, 2019 for reviews). Some accounts propose that the Approximate Number System (ANS) provides the foundation for this process, such that symbolic number words are mapped onto the imprecise numerical representations of the ANS. Meanwhile, others propose that the Object Tracking System (OTS) plays a critical role as children learn to map number words to small quantities within the subitizing range first and only over time do they learn to extend this process to larger numbers.
Here, we asked how young children who have not yet acquired cardinal principle knowledge map number words to the referred quantities in a two-alternative forced choice Point-to-X task (e.g., “Which has three?”). After two practice trials with two different common objects where children were prompted to point to one image (e.g., “Which has a ball?”), they completed sixteen number-word trials. Each trial showed two sets of identical stimuli differing only in number (e.g., four ducks and five ducks), and children were prompted to point to one of the images (e.g., “Which has four ducks?”). Number word trials were designed with varying ratios and sizes of the response options to test whether performance was ratio-dependent, regardless of the size of the presented options (as suggested by the ANS account), or size-dependent, regardless of the ratio (as suggested by the OTS account). 79 children (39 males; Mage=3y9m, range: 3y4m-4y3m) completed the Point-to-X task remotely with an experimenter via Zoom. Overall performance was above chance (Table 1). Critically, performance was better for trials where the ratio between response options was large (e.g., 3 vs. 9) than when the ratio was small (e.g., 2 vs. 3), p < .0001, regardless of the size of the numbers. However, performance was also better for trials where both numbers were small (e.g., 2 vs. 3) than where both numbers were large (e.g., 6 vs. 9), p < .0001, regardless of the ratio between the response options. Thus, children’s performance was modulated by both the ratio and the size of the numbers (Figure 1), suggesting that both the ANS and OTS play a role for children’s number word acquisition.

Authors

• Alex M. Silver, University of Pittsburgh
  Presenting Author

• Leanne Elliott, University of Pittsburgh

• Heather J. Bachman, University of Pittsburgh

• Elizabeth Votruba-Drzal, University of Pittsburgh

• Melissa E Libertus, University of Pittsburgh