Young children develop causal knowledge through everyday family conversations and activities. Children's museums are an informative setting for studying the social context of causal learning because family members engage together in everyday scientific thinking as they play in museums. In this multisite collaborative project, we investigate children's developing causal thinking in the context of family interaction at museum exhibits. We focus on explaining and exploring as two fundamental collaborative processes in parentâchild interaction, investigating how families explain and explore in openâended collaboration at gear exhibits in three children's museums in Providence, RI, San Jose, CA, and Austin, TX. Our main research questions examined (a) how openâended family exploration and explanation relate to one another to form a dynamic for children's learning; (b) how that dynamic differs for families using different interaction styles, and relates to contextual factors such as families' science background, and (c) how that dynamic predicts children's independent causal thinking when given more structured tasks. We summarize findings on exploring, explaining, and parentâchild interaction (PCI) styles. We then present findings on how these measures related to one another, and finally how that dynamic predicts children's causal thinking.
In studying children's exploring we described two types of behaviors of importance for causal thinking: (a) Systematic Exploration: Connecting gears to form a gear machine followed by spinning the gear machine. (b) Resolute Behavior: Problemâsolving behaviors, in which children attempted to connect or spin a particular set of gears, hit an obstacle, and then persisted to succeed (as opposed to moving on to another behavior). Older children engaged in both behaviors more than younger children, and the proportion of these behaviors were correlated with one another.
Parents and children talked to each other while interacting with the exhibits. We coded causal language, as well as other types of utterances. Parents' causal language predicted children's causal language, independent of age. The proportion of parents' causal language also predicted the proportion of children's systematic exploration. Resolute behavior on the part of children did not correlate with parents' causal language, but did correlate with children's own talk about actions and the exhibit.
We next considered who set goals for the play in a more holistic measure of parentâchild interaction style, identifying dyads as parentâdirected, childâdirected, or jointlyâdirected in their interaction with one another. Children in different parentâchild interaction styles engaged in different amounts of systematic exploration and had parents who engaged in different amounts of causal language. Resolute behavior and the language related to children engaging in such troubleshooting, seemed more consistent across the three parentâchild interaction styles.
Using general linear mixed modeling, we considered relations within s...