Lexical networks and their growth in second language learners

What is it about?

Many words in the English language differ by only one sound, such as “bag”, “bat” and “cat”. All the words known by a speaker of a language (the “mental lexicon”) can be charted in a network structure where words are network points and a link is placed between them based on sound similarity, for instance coast, ghost, and toast are linked, and coast is also linked to coat and cost. Mathematical network science offers a way to understand how the structure of the lexical network of second language learners influences the probability of new words to be learned and become embedded in the network of known words in a language user’s mind. The results of this study show that the well-known network growth process of “rich-gets-richer” (or preferential attachment) results in beginning and intermediate language learners preferentially acquiring new words that are similar in sound to many known words. Advanced language learners tend to better learn those words that have fewer similar-sounding neighbor words in the lexical network. This indicates a prevalence of similarity-based word learning at lower proficiency levels of second language acquisition but a tendency to learn dissimilar words at a higher proficiency level. These findings have various implications for lexical learning, memory, and the structure of the second language mental lexicon.

Featured Image

Photo by Alina Grubnyak on Unsplash

Photo by Alina Grubnyak on Unsplash