The Role of Parental Scaffolding in Children’s Referential Pact Formation

• In Event: Broadening the Lens on Parent-Child Interaction as a Mechanism of Developmental Change in Children’s Skills

Abstract

Adults readily form referential pacts (temporary agreement on referent names) with each other when communicating about novel referents (Clark & Wilkes-Gibbs, 1986). On the other hand, children struggle to do so with peers, but the reasons for their failures are less well-studied (Krauss & Glucksberg, 1977). In our study, we explore three possibilities for children’s failures: 1) Children produce overly idiosyncratic expressions; 2) Children are not able to understand the idiosyncratic expressions of others; and 3) Children do not realize that their referring expressions may be ambiguous or under-informative. In 3 experiments, we first ask how parents may scaffold children’s communicative behavior in a reference game. We then turn to examine children’s comprehension and production, to further understand the factors influencing children’s communicative success and failure.

Caregivers may be more sensitive than peers to children’s communicative abilities, and could provide unique support in negotiating referential pacts. In Experiment 1, we ask how parents and children refer to novel objects without conventional labels. Parents and children (ages 4 to 8, n = 61 pairs) played a communication game involving novel tangrams. Analyses reveal that parents and children use shorter sentences as the game progresses, indicating referential pact formation, and replicating Clark and Wilkes-Gibbs’ (1986) results. We find that parents use longer expressions with younger children, and shorter ones with older children, indicating that parents are adapting to their children’s abilities. We also find that parents ask more questions to younger children, suggesting that children may receive scaffolding in the form of clarifying questions or prompting. Finally, we find that referential pacts are more likely to originate from parents, with younger children being more likely to use their parent’s initial descriptions of a tangram in future rounds. Taken together, our results show that children can succeed in forming referential pacts with parent scaffolding.

How much of children’s success in Experiment 1 relied on parent support? In Experiments 2 and 3, we consider two accounts of why children struggle with referential communication when engaging with peers. One possibility is that children generate overly idiosyncratic descriptions of tangrams, and communicative success in our interactive game relied on the parent’s familiarity with their own child. Another possibility is that children are not yet skilled at giving sufficiently informative descriptions of novel objects, and success in the interactive game relied on parents’ prompting or other scaffolding behavior. We test these two possibilities using a comprehension task and a production task. Results from our comprehension task show that even 4-year-old children’s descriptions of tangrams are easily comprehensible by naive listeners (n = 357 adults, 80 children), suggesting that children did not provide overly idiosyncratic descriptions in our interactive game. Results from our production task show that children (n = 60) do not provide sufficient information for identifying a referent, suggesting that their success in our interactive game may rely on contingent feedback from their parents. Results from our three experiments shed light on the role that parent scaffolding plays in supporting children’s communicative competence.

Authors

• Ashley Leung, University of Chicago
  Presenting Author

• Robert D. Hawkins, Princeton University

• Daniel Yurovsky, Carnegie Mellon University