Need for Speed: The Role of Speed of Processing in Novel Word Learning

• In Event: Poster Session 10
  In Poster Session: PS 10 Topic - Language, Communication

Abstract

Infants' large variation in their lexical development has been one of the ongoing puzzles of language acquisition research. Different aspects that impact vocabulary development in early childhood have been identified, among them lexical speed of processing (Fernald & Marchman, 2012). Lexical speed of processing is measured by the reaction time during a looking-while-listening paradigm (Fernald, Zangl, Portillo, & Marchman, 2008). Infants are presented with images of two familiar objects and after a silent viewing period, one of the objects is labelled. When the infant disengages their gaze from the distractor object and shifts their eyes to the target object, their lexical speed of processing can be measured.
Past studies have shown that lexical speed of processing predicts concurrent vocabulary size and subsequent vocabulary growth. Furthermore, it has been linked to parental speech input (Hurtado, Marchman & Fernald, 2008). Marchman and Fernald (2008) have found evidence that lexical speed of processing measured when the infants were 25 months old could predict working memory and IQ at eight years of age. Additionally, Lany (2018) has revealed that speed of processing seems to facilitate novel word learning for infants at 17 months. This supports assumptions made by Fernald and Marchman (2012), who suggested that infants who are able to process familiar words quickly have more resources available to acquire new words. This could causally explain the link between vocabulary size and lexical speed of processing.
However, while the importance of lexical speed of processing has been demonstrated by past research, there is still the open question of the underlying mechanism. Is it related to general processing capability or is it a purely linguistic phenomenon? And if the latter holds true, there is still the question whether it is item-specific or a general property of the whole lexicon.
In order to shed light on these questions, we aim to test 80 17-month-old Dutch infants in a novel word learning study. The aim is to replicate Lany’s (2018) work as well as to investigate the lexical speed of processing of newly learned words. We will train the infants on four novel objects with pseudoword labels. We will then test their knowledge of the novel words in a looking-while-listening paradigm. After a short break, we will also measure infants’ lexical speed of processing, using both pairs of familiar objects and pairs of the novel objects. Additionally, we will assess the infants’ visual reaction time as well as the concurrent vocabulary size using the Dutch CDI.
With the help of Bayesian Linear Mixed Effects Models we aim to disentangle the underlying nature of lexical speed of processing. This project is currently in the pilot stage (the local lab is open at the time of submission) and will be pre-registered, including a detailed analysis plan. Testing will begin in November 2020. Using the pre-registered analyses, (preliminary, depending on the testing progress and development of the pandemic) results will be presented at the conference in April.

Authors

• Julia Egger, Max Planck Institute for Psycholinguistics
  Presenting Author

• Caroline Frances Rowland, Max Planck Institute for Psycholinguistics
  Non-Presenting Author

• Christina Bergmann, Max Planck Institute for Psycholinguistics
  Non-Presenting Author