Optimal Training Systems
Abstract
This report investigates the training implications provided by models built for two domains: learning biology in an online course and learning basic flight maneuvers in an unmanned aerial vehicle simulator. The biology model uses rules to decide on study behavior. The flight maneuver model uses instances of expert behavior to decide on correct flight control actions. Both models are implemented in the ACT-R cognitive architecture. Essentially, the path to optimal training in both these cases involves finding the key domain feature to which learning progress is very sensitive. Based on our results, we would posit that explicitly training on these key features would promote more efficient learning.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 30, 2008
- Accession Number
- ADA481902
Entities
People
- Marsha Lovett
- Michael Matessa
Organizations
- Alion Science and Technology