UC San Diego

Self-directed learning in early childhood often approximates knowledge acquisition in science. Both processes require inquiry (generating informative evidence) and inference (drawing appropriate conclusions from evidence). However, despite being prodigious informal exploratory learners, children typically struggle with inquiry and inference in formal scientific reasoning. This disconnect poses a puzzling problem for researchers who explain self-directed learning as 'intuitive science' and for educators charged with teaching children to engage in scientific thinking.In this dissertation, I aim to resolve this disconnect by providing a more accurate understanding of the intuitive abilities that drive self-directed learning. Chapter 2 examines the early, implicit sensitivity to incomplete information underlying children's preference for exploring uncertainties documented by past research. Chapter 3 provides the first empirical contest between two competing theories of self-directed learning and demonstrates young children's preference to select and ability to learn from informative interventions in their own actions. Chapter 4 reexamines long-standing assumptions about the experimental errors made by self-directed learners in scientific reasoning tasks. Far from misunderstanding the need for or nature of controlled experimentation, the results of this study suggest that learners' approach to these tasks is consistent with the logic of competent causal hypothesis testing. The results of these three studies show that young learners' scientific errors do not stem from either inability or unwillingness to engage in inquiry and inference. Instead, it suggests that self-directed learners' intuitive approach to generating and interpreting evidence is shaped by the complexities and concerns of causal learning. Chapter 5 outlines this novel proposal, explaining how it draws on interventionist causal philosophy and how it allows for reinterpretation of self-directed learners' seemingly uninformative or irrational behaviors. I will argue that the evidence for children's successes in informal exploration and failures in formal experimentation are consistent with approaching inquiry and inference as competent causal learners. Taken together, my dissertation work offers a unified explanation of early self-directed learning and points out a novel path toward better understanding the nature of learning in childhood.