Individual Submission Summary
Share...

Direct link:

Organization helps comprehension: Semantic representations differentiate as a result of experience and support language comprehension

Wed, April 7, 2:45 to 4:15pm EDT (2:45 to 4:15pm EDT), Virtual

Abstract

The acquisition of organized semantic representations – encoding word meanings, facts, and concepts – has long been considered a central aspect of cognitive development. This is in large part because semantic structure is believed to support other cognitive processes (e.g., Fisher, et al., 2015; Medin et al., 1997). Despite this central role, we still know fairly little about how semantic structure is acquired from experience and how it supports other cognitive processes. Across two studies, we examined experience-driven changes of children’s semantic structure in biological domains, and whether individual differences in semantic structure in a domain are related to the ability to comprehend new information in that domain.

If children’s semantic representations change as a result of experience, then concentrated exposure to the features that discriminate items within a domain should increase within-domain semantic differentiation. We tested this hypothesis in Study 1 by examining children’s semantic representations of two biological domains (‘bugs’ and ‘plants’) before and after they completed a week-long enrichment experience that targeted one of the domains. To assess their semantic structure we asked children to organize cards depicting individual word referents such that related things were placed close together; we then used the physical distance between pairs of items to measure their semantic relatedness (Goldstone, 1994). At the beginning of the week children differentiated across the two domains (placing pairs of items that included a ‘bug’ and ‘plant’ farther apart relative to pairs of items that included two ‘bugs’ or two ‘plants’) but there was no evidence of within-domain differentiation. By the end of the week, however, children differentiated items within the domain they experienced (see Figure 1). This pattern – earlier across-domain differentiation followed by more protracted within-domain differentiation – matches cross-sectional results, suggesting that concentrated exposure to a domain accelerated semantic differentiation.

Comprehending language often requires making inferences, and inferences are known to rely on semantic representations. Thus, is it possible that differentiated semantic representations support children’s language comprehension. To test this hypothesis, in Study 2 we measured children’s semantic structure in multiple domains, as well as children’s ability to correctly infer the referent of spoken passages targeting those domains (e.g., inferring that an animal has teeth upon hearing that they “…prey on voles, mice, and other small animals”). Children with more differentiated semantic representations were more likely to correctly infer the referent of the passage, but this effect was modulated by the domain of knowledge (see Figure 2). Exploratory analyses of children’s eye gaze suggested different pathways by which children arrived at the correct response in the language comprehension task, suggesting that semantic structure may mediate specific aspects of language-based inferential comprehension.

Together, these studies provide evidence that experience drives changes in semantic structure, and that semantic structure supports a key cognitive process – learning from linguistic input. We discuss the implication of these results for conceptualizing the contributions of enrichment experiences to academic success.

Authors