Poster #65 - Relational abstraction in early childhood: Three contexts and three trajectories

• In Event: Poster Session 19

Abstract

Abstract reasoning in early childhood is often described as following a "relational shift," over which children become less focused on object properties and increasingly sensitive to relations. However, recent work has challenged the generality of this account, showing that children in the US and mainland China follow two distinct trajectories in a causal relational match-to-sample (cRMTS) task between 18 and 48 months (Carstensen et al., 2019). Specifically, children in the US follow a U-shaped learning trajectory, with good performance in relational matching between 18 and 30 months, poor performance between 36 and 48 months, and later recovery. In contrast, children in China show good performance at the youngest end of this age range and no evidence of decline, showing comparable or improved performance by the preschool age. This difference aligns with multiple linguistic and cultural factors implicated in relational reasoning, in which English speakers in the US and Mandarin speakers in China appear at opposite ends of a continuum spanning from a focus on objects (US) to one on relations (CN). In this study, we explore early relational reasoning in a context that represents a cultural middleground with linguistic similarity to the US: Korean-speaking children in South Korea.

This data informs discussions of a key potential mechanism in the trajectory of early relational reasoning: language. Hoyos, Shao, and Gentner (2016) proposed that the overrepresentation of nouns in (English) language learning creates a temporary bias toward objects and object properties, like shape, causing the observed decline in relational matching between toddlers and preschool-aged children in the US. Mandarin Chinese learners, unlike English-learning children, do not show a strong noun bias in their early lexicons (Frank et al., 2021), and also do not show the decline in relational matching that Hoyos et al. attribute to avid noun learning. Korean word learning shows a similar noun bias to that in English (ibid), but provides a cultural middleground between the US and China; if lexical bias alone drives the trajectory of relational matching, then learners in Korea should show performance resembling their peers in the US.

We find that toddlers in Korea are similar to their counterparts in China, showing early success in relational matching in the same condition (n=41, Bayes Factor10=4.07), but not in the different condition (n=50, BF10=.33). Our data suggest that at 3 years old, performance in the Korean sample is intermediate to that in the US and China (same n=50, BF10=.58; different n=50, BF10=.81). Three-year-olds in Korea also show intermediate scores in a bias assessment (n=55, BF10=.34), with no clear preference for the object bias observed in the US or the relational bias observed in China. In line with previous findings, we document a link between culturally variable cognitive preferences and variation in relational matching. At the same time, the current findings underscore variation in early abstract reasoning; our findings suggest that there is neither a single common trajectory nor two typical trajectories in early relational responding, and that both linguistic and cultural factors may contribute to shaping early abilities.

Authors

• Alexandra Carstensen, University of California - San Diego
  Presenting Author

• Minju Kim, University of California - San Diego

• Gayoung Kim

• Meizi Jin

• Minjin Kang

• Caren M. Walker, University of California - San Diego

• Youngon Choi, Chung-Ang University