Gray, W. D., Schoelles, M. J., & Sims, C. R. (2004). Learning to choose the most effective strategy: Explorations in expected value. In M. C. Lovett, C. D. Schunn, C. Lebiere & P. Munro (Eds.), 6th International Conference on Cognitive Modeling–ICCM2004. Pittsburgh, PA.
Learning to choose the most effective strategy: Explorations in expected value
Small variations in how a task is designed can lead humans to tradeoff interaction-intensive for memory-intensive strategies. In this paper we introduce one such task, Blocks World, and present empirical data that shows such tradeoffs. Our attempts to model the acquisition of these tradeoffs using the default ACT-R conflict resolution mechanisms have met with failure. We have now run our model using four different methods for tallying a strategy’s success and failure. For each run, we discuss the formulation of expected value used for conflict resolution and the reasons why the model does or does not match the human data.Download Now Back to Home << Publications Visitors since 2004.12.08: