Research Projects

Explanation & Causation

The process of seeking, generating, and evaluating explanations has been proposed to serve as a powerful mechanism underlying children's learning, scaffolding knowledge acquisition and contributing to theory change. We are investigating the precise role of explanation in young children's causal learning, focusing on whether and how explaining influences the process of early reasoning and inductive inference.

• Does explaining influence the relative contributions of observed evidence vs. currently-held beliefs in children’s learning and generalization?

• Can explaining prompt children to privilege hidden, but inductively-rich properties over salient surface similarities?

• Does explaining promote a bias to prefer simple hypotheses over complex ones?

• Does explaining facilitate abstraction and foster the acquisition of abstract concepts?

• How can we better integrate empirical findings on the role of explanation on early learning with theoretical perspectives from computational modeling?

• Does explanation impact learning by changing the hypotheses that children initially generate in response to a problem?

• What is the relationship between explanation and understanding? Must an explanation be true in order to invoke understanding?

• Can prompting children for multiple explanations support recognition of uncertainty?

• What is the role of pedagogical context in mediating the effects of explanation?

Intuitive Science and Self-Directed Learning

Decades of research in cognitive development suggests that learners are “intuitive scientists,” generating informative actions and rationally coordinating their observations and prior beliefs. At the same time however, research on the development of formal scientific reasoning finds learners struggle to conduct informative experiments or arrive at valid conclusions. The prevailing narrative in both disciplines has been to explain away, or ‘bridge,’ this apparent gap between children’s informal competence and formal failures. In our lab, we advocate for an alternative approach that interprets self-directed learning in terms of the particular goals and considerations of causal reasoning. In doing so, we aim to provide a more complete and accurate account of self-directed learning in childhood.

• Are there alternative explanations for the classic errors of scientific inquiry and inference documented in childhood and beyond?

• Is early exploration driven by children’s desire to produce tangible outcomes or information?

• Are children able to learn from the information they generate themselves during self-directed exploration?

• How might elements of task design impact children’s ability to conduct experiments and evaluate evidence (e.g., in control of variables tasks)?

• Can learners’ failures to use a falsificationist approach to hypothesis testing (e.g., disconfirming alternatives) be reconceived as a rational and informative approach to inquiry?

• Do learners seek information about causal invariance (i.e., how reliably a causal relationship holds within and across contexts), and does this explain their tendency to generate positive hypothesis tests?

• Do children’s beliefs about the variability of their environment impact their exploration strategies?

Relational Reasoning

• When can children learn the abstract concepts same and different? What about other higher-order principles, like variability and uniformity? Can they use these relations to make predictions about novel outcomes they have never observed?

• Does children’s early success on relational reasoning tasks result from genuine conceptual understanding or reliance on perceptual heuristics and low-level cues?

• Can young children generalize abstract relations across distinct dimensions and modalities? How robust are these early representations?

• Do learned biases impact children’s abstract reasoning abilities? How might these biases be influenced by language and culture? Are there spontaneous differences in the developmental trajectory of relational reasoning across cultures?

• What factors in the learning environment facilitate children’s recognition of abstract relations? Can relational reasoning be primed? Can the design of physical objects in the environment support children’s attention to abstract relations?

• Can providing prompts to explain or gesture encourage generalization of abstract categories and problem solutions?

• Do children recognize causal system categories, or abstract patterns of causation that apply across phenomena (e.g., causal chain, positive feedback loop)? Can this recognition be used to facilitate knowledge transfer in educational contexts?

Understanding Uncertainty

Understanding uncertainty is essential for learning, and this uncertainty appears in a variety of forms. First, there is uncertainty regarding outcomes in the world. Since we rarely have complete information about causal structure, we are forced to consider which of several alternatives is most likely, invoking uncertainty. Uncertainty also arises whenever we predict the future, since more than one possible outcome is usually available. Finally, we often express subjective uncertainty about the accuracy of our own knowledge. What do children understand about the presence of uncertainty in the world and/or in their own knowledge? When and how does this understanding develop?

Thinking Counterfactually

This line of work explores the relationship between children's understanding of real-world causal structure and their engagement with alternative possibilities, hypothetical reasoning, and pretend play. This research is broken into several distinct lines 1) Can counterfactual reasoning be used to support the development of skills related to scientific inquiry? 2) Is there a relationship between childhood pretense and causal counterfactual reasoning?

Fictional Cognition and Thought Experiments

Given the amount of information that is presented to children in the context of fictional stories and imaginary representations of the world, it is important to understand the inferences that underlie children’s causal learning in this domain. More broadly, how and why do we learn about the real world from things that are false? In particular, under what conditions do we learn even better from falsehoods than from truth? This happens when we read fiction, when children engage in pretend play, when philosophers reason from thought experiments, and when scientists reason from idealized models or simulations. Why are these falsehoods so helpful and so central to human reasoning?

• Can children’s literature be used to promote children’s understanding about the nature of knowledge?

• Can children’s literature be used to promote children’s argumentation skills?

• Does the presence of fantastical content influence children’s ability to learn novel causal information from storybooks?

• What do children understand about the boundary between the fictional and real world?

• Can exposure to fictional stories prompt conceptual change about physics?

• Why does engaging in fictional (e.g., thought experiments) or idealized worlds (e.g., models in science) facilitate learning and understanding in adults?