The company that words keep: comparing the statistical structure of child- versus adult-directed language

Abstract

Original citation: Hills, Thomas Trenholm. (2013) The company that words keep: comparing the statistical structure of child-versus adult-directed language.

Semantic facilitation in bilingual first language acquisition

“…This indicates that, among other influences-such as phonological neighborhoods, word repetition, and pragmatics (Lust, 2006;Storkel, 2004;Tomasello, 2009)-one of the key influences on early word learning is the statistical structure of the language learning environment. Consistent with this claim is the observation that adults amplify the associative structure of their language when speaking to children, specifically producing more associates (as produced by adults in free association norms) around words that children learn earliest (Hills, 2012). In addition, several studies have now established that children during their earliest years of language learning are influenced by the semantic (i.e., associative) relations between words (Arias-Trejo & Plunkett, 2013; Delle Luche, Durrant, Floccia, & Plunkett, 2014;Styles & Plunkett, 2009;Willits, Wojcik, Seidenberg, & Saffran, 2013).…”

“…Both groups show a clear affinity for learning hubs in the free association norms and follow a pattern of acquiring high associative indegree words early on. This is potentially a result of the architecture of child-directed speech, where the probability of producing associates around early learned words is nearly double that in adult-directed speech-thereby amplifying the semantic relations most around the words that children learn earliest (Hills, 2012). Associative structure may also be correlated with other properties of child-directed speech likely to influence early learning-such as repetitions, contextual diversity, and frequency (e.g., Hills, 2012;Hills et al, 2010)-and many other factors besides language structure are known to influence language acquisition (Lust, 2006;Tomasello, 2009).…”

“…For monolinguals, a word's age of acquisition is correlated with its associative relationships with other words (Hills, 2012;Hills et al, 2009b). If bilingual children show the same pattern, then their earliest learned English words should also show a bias for higher associative indegree.…”

“…In this section, we examine two versions of these learning rules: preferential acquisition and preferential attachment. For preferential acquisition, words are learned in relation to their associative indegree, a relationship that is known to be more salient in the early learning environment as a result of changes in child-directed speech (Hills, 2012). An example of preferential acquisition would be learning the word 'ball' more quickly than 'telephone' because 'ball' is a word that is produced more often as a cue in free association norms.…”

Estudio del lenguaje desde la perspectiva sociocultural

“…Por ejemplo, en Hills (2012), se comparó la estructura del lenguaje dirigido a niños y adultos en su modo de asociación o repetición como método de aprendizaje de palabras. En el mismo, se demostró que el lenguaje se estructura de manera diferente cuando se dirige a aprendices del mismo que cuando se dirige a hablantes de alto conocimiento del idioma, específicamente en formas que aparentan estar correlacionadas con el aprendizaje temprano del lenguaje como: estructura asociativa, diversidad contextual, repetición y frecuencia del uso de las palabras.…”