Poster #72 - The dynamics of child-driven active information-seeking and caregiver scaffolding during word learning

• In Event: Poster Session 01

Abstract

Introduction.
Caregiver-child dyadic interactions are central to early word learning and set the trajectory for later language development (Romeo et al., 2018). Early language learning emerges from dynamic communicative exchanges characterized by two key dimensions. First, children are not simply passive observers, but instead actively seek new information. Second, caregivers are attuned to children’s learning needs and scaffold the learning environment. While each of these dimensions have often been studied in isolation, they are interdependent: what information-seeking strategies are most beneficial depends on caregivers’ responses; what scaffolding is most beneficial to children depends on their information-seeking strategies.

The current study.
This study aims to determine how caregiver scaffolding and children’s information-seeking mutually contribute to effective word learning. We will test this question in a novel word learning paradigm that independently varies whether children and/or their caregivers can structure the learning input in a joint Learning Phase, and measures the increase in children’s word knowledge in a pre-post test design.

Design.
We will test 96 children (36-72 months) and their caregivers in a word learning task. Children will complete 4 experimental rounds. In each round, they will learn and be tested on their knowledge of 8 animal labels ranging from very well known (e.g., butterfly) to unknown (e.g., echidna), as predicted by age-of-acquisition norms. Each round consists of a Pre-Test Phase, a Learning Phase and a Post-Test Phase. In the pre-test, children are tested on their word recognition accuracy in an 8-alternative forced-choice task.

Then, each iteration of the Learning Phase has two steps (Fig. 1). The first step is a Scaffolding event, in which two object-label candidates are selected from a set of 8 object-label pairs. In the Caregiver Scaffolding condition, the caregiver selects two object-label pairs for children from among the 8 object-label pairs (Fig. 2). In the Random Scaffolding condition, the two object-label pairs are selected at random. In the second step, children hear the word for one of the two object-pairs in one of two Sampling events (crossed with the scaffolding condition). In the Active Sampling condition, children choose which word they hear next. In the Random Sampling condition, the object-label pair the child hears will be selected at random. In the Post-Test Phase, children complete the same word recognition task as in the pre-test, in order to measure the increase in children’s word recognition accuracy.

Predictions.
We have two main predictions
(1) When given the opportunity to structure the Learning Phase, both children and caregivers will systematically structure the learning events to help children learn less well-known words.
(2) Children will show learning benefits when they themselves and/or their caregivers can structure the Learning Phase, relative to the condition in which the learning curriculum is generated randomly.
We will fit mixed-effects models to test each of these predictions (project page: https://osf.io/p8mcs/?view_only=ea3790d547a9473283d6acbbc347e76c; preregistration: https://osf.io/s24eb?view_only=7cbed0ef12354103b5fe7a897d13f694).

Conclusions.
The current experiment will provide insight into how children’s information-seeking strategies interact with caregivers’ scaffolding strategies to create informative word learning moments.

Authors

• Martin Zettersten, Princeton University
  Presenting Author

• Casey Lew-Williams, Princeton University