作者: Richard Seheines , Matt Easterday , David Danks
DOI: 10.1093/ACPROF:OSO/9780195176803.003.0009
关键词:
摘要: There is now substantial agreement about the representational component of a normative theory causal reasoning: Causal Bayes Nets. less discovery from data, either computationally or cognitively, and almost no work investigating how teaching Nets apparatus might help individuals faced with learning task. Psychologists working to describe naive participants represent learn structure data have focused primarily on single trials under variety conditions. In contrast, one focuses sample drawn population some experimental observational study regime. Through virtual Causality Lab that embodies reasoning which allows us record student behavior, we begun systematically explore best teach theory. this paper explain overall project report pilot studies suggest students can quickly be taught (appear to) quite rational. Acknowledgements We thank Adrian Tang Greg Price for invaluable programming Lab, Clark Glymour forcing get point, Dave Sobel Steve Sloman several helpful discussions. * This research was supported by James S. McDonnell Foundation, Institute Education Science, William Flora Hewlett National Aeronautics Space Administration, Office Naval Research (grant Human Machine Cognition: Systems Technology Address Critical Navy Need Present Future 2004). 1 Dept. Philosophy Human-Computer Interaction at Carnegie Mellon University. 2 Mellon. 3 Department Philosophy, Mellon, Cognition, University West Florida.