Developmental Psychology Colloquium

Colloquium | April 15 | 12:10-1:15 p.m. | 1102 Berkeley Way West

 Katie Kimura, Department of Psychology; Yuan Meng, Department of Psychology

 Department of Psychology

Katie Kimura will present her graduate research:
Title: Learning and then re-learning: Belief revision in young children
Learning is a complex process that requires integrating new evidence with existing beliefs. This task is relatively straightforward when the evidence is consistent with and thus supported by existing beliefs. It becomes less clear, however, when the evidence is inconsistent—that is, when the evidence directly conflicts with one’s prior knowledge. Will learners revise their beliefs to reflect the new evidence or will they instead maintain their existing beliefs, perhaps ignoring the conflicting data as anomalous? In this talk, I will present two studies that being to address this question—one completed and one ongoing—that demonstrate young children’s remarkable ability to normatively revise their higher-order beliefs by attending to both the weight of their initial belief and the strength of the counterevidence. Taken together, these studies reveal how children are rational revisionists of their abstract, higher-order beliefs.

Yuan Meng will present her graduate research:

Title: Leveraging Thinking to Scaffold Causal Learning from Intervention

Abstract: Learning the right causal structure from intervention asks that the learner choose the right intervention and make the right inference based on interventional data. In the current study, we sought to scaffold both processes by asking 5- to 7-year-old children to explain why they planned to carry out certain intervention in order to identify the true structure of unknown causal systems that might work in one of two ways. Compared to the control group who only reported which intervention to choose, explainers did not shift towards the optimal strategy, i.e., maximizing the expected information gain (EIG) of the chosen intervention. However, the scaffolding we provided—be it explaining or reporting intervention choices—increased children's accuracy at inferring the true structures, potentially by helping them better utilize interventional data that they generated on their own.

 lisabranum@berkeley.edu