Child Development Perspectives | 2019
Learning to Remember Words: Memory Constraints as Double‐Edged Sword Mechanisms of Language Development
Abstract
Word learning researchers have historically been interested in elucidating the mechanisms that allow children to encode words. Recent research has moved beyond the moment of encoding to examine the processes underlying children’s retention and retrieval of words across time. This work has revealed significant memory constraints on children’s word learning. That is, children struggle to retain and retrieve newly learned words. This review outlines research suggesting that describing these processes as memory constraints may mischaracterize how memory shapes language development. Instead, memory constraints are more accurately characterized as double-edged sword mechanisms; limited memory abilities likely hinder and promote children’s word learning simultaneously. The review concludes with suggestions for developing a theory of how children learn to remember words. KEYWORDS—forgetting; retrieval effort; word learning; memory development; language development; cognitive development Linguists, philosophers, and psychologists have historically characterized language learning as seemingly impossible (e.g., Quine, 1960). Word learning is particularly difficult because children need to resolve high degrees of referential ambiguity to determine word-referent pairings. For instance, if children are at the park and hear the word dog, this word could refer to the creature they see wagging its tail and barking, but it could also refer to a seemingly infinite number of other referents, such as a color, action, and abstract concept. Thus, a central pursuit in cognitive science, developmental psychology, and other disciplines has been to understand the mechanisms that make word learning possible. Over the last few decades, researchers have focused on studying how children learn to associate words with referents in a single moment. In these studies, children are exposed to a novel word and a novel object, then are immediately tested on whether they mapped the word to the correct object. For instance, in a typical experiment (see Figure 1), children are presented with a novel linguistic label (wug) and a novel object. At test, children are shown several objects and asked to identify the wug. If children recognize or generalize the word wug at an immediate test, researchers conclude that children successfully encoded the word presented to them. Indeed, researchers have found that children can readily map words to referents after just a few learning trials, a behavior termed fast mapping (see Carey, 2010, for a review). Three classes of theories have been proposed to explain children’s rapid mapping of words to referents: domain-general theories (Smith, 2002), constraints-principles theories (Markman, 1991), and social-pragmatic theories (Tomasello, 1992). Taken together, these theories suggest that children use a variety of tools to determine word mappings. For instance, children use associative/statistical learning to track regularities in perceptual information (Krogh, Vlach, & Johnson, 2013), apply rules (e.g., mutual exclusivity; Kalashnikova, Mattock, & Monaghan, 2016), and rely on the social cues of others (Yurovsky & Frank, 2017) Haley A. Vlach, Department of Educational Psychology, University of Wisconsin, Madison. I thank my colleagues for their insights, feedback, and continual support over the years. Special thanks to Megan Kaul and Melina Knabe for reviewing this article. The process of writing this article was supported in part by National Institutes of Health grant R21 HD092867-01. Correspondence concerning this article should be addressed to Haley A. Vlach, Department of Educational Psychology, 859 Educational Sciences, 1025 W. Johnson Street, Madison, WI 53706; e-mail: [email protected]. © 2019 Society for Research in Child Development DOI: 10.1111/cdep.12337 Volume 13, Number 3, 2019, Pages 159–165 CHILD DEVELOPMENT PERSPECTIVES