Poster #13 - Why do Children Acquire Small Number Words in Order?

• In Event: Poster Session 06

Abstract

Universally across modern cultures, children acquire the meaning of the number words one, two, and three in the order of their magnitudes (Le Corre & Carey, 2007; Wynn 1990, 1992). Most research has focused on how children acquire this knowledge (i.e., the general acquisition process) or what this knowledge represents (i.e., the limitations of this knowledge), while the question of why children learn numbers in order has been comparatively neglected. In other words, why do children always learn the number one first? The present work contributes to this scientific inquiry by asking whether preverbal infants represent and relate sets of one differently than sets of two or three. To address this question, a non-verbal categorization paradigm called implicit concept formation was imported from the language acquisition field. (Shukla & de Villiers, 2021).

This paradigm was implemented across three asynchronous online experiments hosted on Lookit. Infants (14-22 months) observed a target stimulus and a non-target stimulus presented side by side on a monitor. The two stimuli were first gray scaled and static for a short time called the anticipatory period. Following this period only the target illuminated and flashed bright colors with an accompanying reward sound. Over many trials the infants’ ability to form the concept implied by the target stimulus was assessed by quantifying the ratio of looking time towards target versus non-target during the anticipatory period. Importantly, this paradigm evaluates the context independence of relational representations by requiring children to relate previously seen stimuli to new stimuli across occlusions of time and space. Three number conditions were administered across the experiments, where either one, two, and three homogenous shapes served as the target with the other two numbers used as the non-target. Similarly, three control conditions were administered where three different line orientations served as target and non-target. The control condition was included to ensure that the implicit concept formation task could be transferred from a controlled lab setting to the home environment.

Our hypotheses are twofold. First, preverbal infants might better represent a single item in a relational manner than two or three items, leading to the pattern of number acquisition previously described. If true, one would predict that the infants will form the target concept of one more readily than the target concept of two or three across the experiments. Alternatively, preverbal infants might require linguistic training before representations of small set quantities are naturally used in a relational manner; therefore, the order of acquisition would be driven by cultural input. This hypothesis leads to the prediction that infants should form the target concepts of one, two, and three in a comparable manner. These hypotheses will be tested by statistically comparing the infants’ average target versus non-target looking time ratios across the three experiments. This analysis will allow us to determine whether the numbers one, two, and three are represented and related differently from each other. Data from 14 infants has already been collected but we are targeting a sample size of 60 participants.

Authors

• Jenna Croteau, University of Massachusetts Amherst
  Presenting Author

• Joonkoo Park