Poster #110 - Consistent exposure to word repetition supports two-year-olds’ further learning from repetition

• In Event: 3-206 - Poster Session 13
  In Poster Session: PS 13 Section - Language, Communication

Integrative Statement

Variation in the quality and quantity of child-directed speech (CDS) shapes young children’s vocabulary growth (e.g., Hirsh-Pasek et al., 2015; Huttonlocher et al., 2010; Weisleder & Fernald, 2013). One feature of CDS that supports learning is parents’ repetition of words across time (e.g., Newport et al., 1977). Schwab and Lew-Williams (2016) revealed that young children best learn novel words when words are repeated in neighboring sentences. However, the presence of useful cues for learning, including repetition, varies across different children’s language environments (Newman, Rowe, & Ratner, 2016; Schwab, Rowe, Cabrera, Lew-Williams, 2018). It is not currently known how the ability to benefit from such cues may build on itself with more vs. less practice. Children may best capitalize on repetition when it is a consistent cue; alternatively, children may also best capitalize on repetition when it is a relatively rare cue – thereby drawing children’s attention to structure within an otherwise confusing environment (e.g., Kidd et al., 2012). Here, we simulated the presence vs. absence of repetition as a consistent cue in order to understand how surrounding the language environment can best be packaged to support learning. Specifically, we asked how exposure to repetitions of familiar words affects learning of repetitions of novel words.

Method. We used a two-phase looking-while-listening procedure to teach toddlers two novel words. English-learning two-year-olds (n=38) were randomly assigned to one of two conditions: “Repetition Context” or “Non-Repetition Context”. In the learning phase, participants in both conditions heard each of two novel words repeated in three successive trials. Critically, these blocks of novel-word trials were preceded by two blocks labeling familiar objects (Fig. 1). In the Repetition Context condition, familiar-word blocks consisted of object labels repeated across three successive trials. In the Non-Repetition Context condition, participants saw the exact same set of familiar trials, but semi-randomized such that there was no immediate repetition of familiar items. On each of 12 test trials, children viewed the two novel objects side-by-side while listening to a sentence referring to one object (e.g., “Where is the fep?”).

Results. Accuracy in looking to the target picture was measured for each condition. Accuracy differed significantly from chance in the Repetition Context condition [M=.59; t(19)=2.67, p=0.02], but not in the Non-Repetition Context condition [M=.53; t(17)=.92, p=0.39] (see Fig. 2). Thus, toddlers successfully demonstrated learning when they heard novel words presented in a repetition context, but not when they heard these words following a context with no repetition.

Discussion. These findings suggest that consistent exposure to a useful cue, such as repetition, facilitates young children’s abilities to capitalize on that cue when learning new information. Contexts containing regular use of repetition may provide young children with expectations about incoming information, supporting their ability to exploit this naturally-occurring feature of CDS. Therefore, not only are characteristics of CDS useful in isolation, but they may also be useful in setting the stage for future learning. Findings will be discussed in the context of dynamic systems theories of development, as well as broader research on ‘learning to learn’.

Authors

• Eva Fourakis, Princeton University
  Presenting Author

• Christine Potter, Princeton University
  Non-Presenting Author

• Jessica Schwab, Harvard University
  Non-Presenting Author

• Casey Lew-Williams, Princeton University
  Non-Presenting Author