Leveraging Thinking to Facilitate Causal Learning from Intervention
- Yuan Meng, Psychology, University of California, Berkeley, Berkeley, California, United States
- Fei Xu, Psychology, UC Berkeley, Berkeley, California, United States
AbstractIntervention selection is at once crucial in causal learning and challenging for causal learners. While the optimal strategy is maximizing the expected information gain (EIG), both children and adults often combine it with suboptimal ones such as the positive test strategy (PTS). In the current study, we sought to facilitate causal learning from intervention by asking 5- to 7-year-olds to explain why they chose a certain intervention to identify the true structure of a three-node causal system that might work in one of two ways. Our findings suggest that while engaging in self-explaining did not help children select more informative interventions, asking them to think about their intervention choices (explaining or reporting) might help them better utilize interventional data to infer causal structures.
Return to previous page