“If you're a 5 year old and you go to the grocery store with your mom or dad, it’s an opportunity for learning. Any time parents and children interact outside of school, it's a chance for children to learn,” said Dave Sobel, Professor of Cognitive and Psychological Studies and former Director of Undergraduate Studies.

Sobel and his lab, the Causality and Mind Lab, has been studying the development of scientific reasoning and children’s engagement with science across different locations, mediums, and ages. Sobel and his lab have recently published in journals like Child Development, Developmental Science, Monographs of the Society for Research in Child Development, as well as the book “Constructing Science: Connecting causal reasoning to scientific thinking in young children.”

Sobel created the Mind Lab, a program of exhibits, science education, and child participation, in collaboration with the Providence Children's Museum in 2010. Working with a team of researchers in cognitive development and practitioners in museum education, he studied parent-child interactions, how they “played together at museum exhibits and what they learned and what they learned about learning,” as Sobel explained. Between the Providence Children's Museum, Children’s Discovery Museum (California), and The Thinkery (Texas), they ran a decade-long nationwide investigation of parent-child interaction. Their research found that parental behaviors at the exhibit affected child learning and engagement, Sobel shared, but that there were two distinct sets of behaviors for each of the two factors.

Sobel explained, “So what we've been developing is the idea that there are things that parents do that support children's autonomy in a learning environment. And the key point is that anything is a learning environment.”

His research, he emphasized, at museums and in terms of parent-child interactions has centered on science, learning scientific material, and utilizing scientific reasoning.

Sobel and his collaborators also sought to pursue connections between play and learning, emphasizing that children understand the relation between the two concepts before they enter formal education.

“For me,” Sobel added, “[it is] an incredibly important finding because it's not [that] you can't have any fun when you're learning. What you really are seeing is that children actually quite naturally integrate those two things [play and learning] together. Now, that takes some time to develop…. And I do think that the way we talk about these concepts can sometimes be a little challenging for kids. Sometimes you talk about play and learning as separate things. But the more children are exposed to the idea that there's playful learning, I think the better they're going to be in terms of how they engage in those activities.”

Another of Sobel’s studies questioned children (five-to-eight-year-olds) in their ability to make causal inferences about science. Students were challenged to identify the source of the results in confounded and unconfounded experiments and asked to answer open-ended questions about their understanding of science. Although all children had the ability to make these inferences, those who had stronger conceptualizations of science—knowledge “that everyday actions are scientific and that science is a general problem-solving process”—performed better on the task, Sobel shared. This finding controlled for age and linguistic ability.

Notably, the causal reasoning tasks involve the Blicket Detector, a device Sobel helped develop as a graduate student in 1995.

These results echoed another longitudinal study of Sobel’s, where the performance of Pennsylvania elementary school students on causal reasoning tasks in third grade was correlated with their performance on the statewide science standardized test (the Pennsylvania System of School Assessment) a year later.

Multiple factors must be considered when assessing performance. “Science is hard,” Sobel acknowledged. “One of the things that we've actually shown is that one of the things that makes science hard is all of the terminology.”

Although children can make decontextualized inferences, complicated terminology hindered their ability to act, he shared. 

As an example, he reflected, “Kids love dinosaurs. But once you make the problem about dinosaurs, all of a sudden 10-year-olds can’t do the kinds of inferences that a 4 and 5 year old could do. Dinosaurs have complicated names and complicated features, and having to remember all of that information can get in the way of making inferences. Even though the inference we are asking them to make is exactly the same, the complexity of the material reduces the child's ability to engage in that inference.”

This inference ability, Sobel argues, comes both from internal mechanisms and social factors that affect child autonomy.

“One of the things that we see over and over again is when parents reduce their support for children’s autonomy, when they engage in behaviors that prevent children from doing things on their own,” Sobel explained. “Those children are less engaged, they play less for less amount of time, and they're less willing to challenge themselves when things get difficult.”

In exploratory play environments, his team found, “the children who have parents who collaborate with them and the children who have parents who are more controlling usually learn equivalently. If you're a parent and your only goal is to have your kid learn, collaboration doesn't matter, but if you want your kid to be engaged as an active learner, then autonomy support matters a great deal.”

Furthermore, parents who provided more causal talk (e.g., asked why-questions or generated more causal utterances had children who learned more from their play.

One of his most recent papers even found that children in collaboration with their parents (where parents direct them) are less willing to attempt the harder challenges than those with no parental involvement.

“It really suggests there's something about reducing that autonomy support that's lowering kids engagement, not that being with your parent raises the engagement,” Sobel added.

In some of his newest work, supported by the National Science Foundation, Sarah Kiefer, a graduate student in the Causality and Mind Lab, ran an online study on parent-child interaction in the home. Children and parents participated in a building task together with a maker kit that the lab sent them. Dyads built a train together out of paper, which focused on connecting the cars together in a specific spatial configuration. Afterwards, children completed a set of challenges about spatial reasoning, which got progressively more difficult. Kiefer found that parent-directed children (those who were instructed by parents rather than encouraged during the construction) were less willing to participate in the challenges, compared to children whose parents were more collaborative or hands-off, or children who did not participate in this interaction.

Across the many studies Sobel has published recently, findings emphasized the importance of parental backing, accessible language, and developing an understanding of science. “The goal of many of the studies I do is to try to foster a more scientifically literate society. This starts with encouraging children to engage with science,” Sobel shared.