Categorical structure among shared features in networks of early-learned nouns

Abstract

The shared-features that characterize the noun categories that young children learn first are a formative basis of the human category system. To investigate the potential categorical information contained in the features of early-learned nouns, we examine the graph-theoretic properties of noun-feature networks. The networks are built from the overlap of words normatively acquired by children prior to 2 ½ years of age and perceptual and conceptual (functional) features acquired from adult feature generation norms. The resulting networks have small-world structure, indicative of a high degree of feature overlap in local clusters. However, perceptual features—due to their abundance and redundancy—generate networks more robust to feature omissions, while conceptual features are more discriminating and, per feature, offer more categorical information than perceptual features. Using a network specific cluster identification algorithm (the clique percolation method) we also show that shared features among these early learned nouns create higher-order groupings common to adult taxonomic designations. Again, perceptual and conceptual features play distinct roles among different categories, typically with perceptual features being more inclusive and conceptual features being more exclusive of category memberships. The results offer new and testable hypotheses about the role of shared features in human category knowledge.

Development of inductive generalization with familiar categories

“…Recall that early in development, semantic representations are poorly differentiated but become increasingly differentiated in accordance with taxonomic relations in the course of learning and development Hills, et al, 2009;Kemp & Tenenbaum 2008;Rogers & McClelland, 2004;Unger et al, 2014). For example, Unger et al (2014) observed that preschoolers and kindergarteners were likely to group together items, such as whale, seaweed, and water lily (likely due to similarity in habitat); however, by second grade children grouped items largely in accordance with the plants-animals distinction.…”

“…In contrast, when reasoning about novel artificial categories, children's inferences cannot be category-consistent in their outcome without being categorybased in their mechanism. Newly learned artificial categories are based on a small number of shared features and their representations are likely not sufficiently differentiated to support inferences based on representational similarity (Hills, et al, 2009;Kemp & Tenenbaum 2008;Rogers & McClelland, 2004). Therefore, inconsistencies in the developmental trajectory of inductive generalization that have emerged across different studies can be explained by the notion of representational similarity.…”

“…Computational models of semantic development suggest that representational similarity undergoes dramatic changes between the preschool and the early school years (Hills, Maouene, Maouene, Sheya, & Smith, 2009;Kemp & Tenenbaum 2008;Lowe, 1997;Rogers & McClelland, 2004). Specifically, concepts are organized early in development on the basis of a small number of salient features that support the formation of relatively undifferentiated groups.…”

“…Semantic organization that includes a single densely interconnected cluster (Hills et al, 2009) or several poorly differentiated clusters (Kemp & Tenenbaum, 2008;Rogers & McClelland, 2004) is arguably ill suited for supporting categoryconsistent inferences. For instance, if one has to predict whether a crocodile is more likely to share an internal property with an alligator or a penguin, this task is trivial if representational similarity between the concepts of crocodile and alligator is higher than between the concepts of crocodile and penguin.…”

Quantifying the structure of free association networks across the life span.

“…One of the advantages of network analysis is that it allows researchers to study phenomena at scales that range from individual words (i.e., nodes) to the entire lexicon (i.e., the network). Network analyses have been widely used to study the large scale structures of associative networks (De Deyne, Navarro, & Storms, 2013;De Deyne & Storms, 2008a;Steyvers & Tenenbaum, 2005), the formation of categories in toddlers (Hills, Maouene, Maouene, Sheya, & Smith, 2009a), and the trajectories of network development during early life (Bilson, Yoshida, Tran, Woods, & Hills, 2015;Hills et al, 2009b).…”