Funded by the Office of Naval Research
Cognitive and Neural Sciences Division
Program Officer: Dr. Susan Chipman
Award Number N000140210152
2004-2007, $923,971.00

PI: Dr. Stephanie Doane
Team Members: Dr. Gary Bradshaw, Ben Craig, Andrew Egerton, Dave Wilson

Abstract:
The present research uses a cognitive theory-driven approach to explore training opportunities in ADAPT, a Construction-Integration based cognitive model that has been transitioned to serve as a real-time intelligent tool for tutoring instrument flight skills. The ADAPT model has been demonstrated capable of predicting individual novice, intermediate, and expert pilot simulated instrument flight performance (Doane & Sohn, 2000). In addition, ADAPT has learning mechanisms to acquire both procedural and declarative knowledge. The mechanisms have demonstrated capable of predicting individual student learning from a computer-based tutor for computer command skills (Doane & Sohn, 2000; Sohn & Doane, 2002). For purposes of experimental tractability, a synthetic task (simulated instrument flight) has been chosen. This task preserves much of the structure of a real-world task like instrument flight but is simplified to enable us to complete this research with a relatively small budget. Previous ONR funds were used to modify ADAPT to facilitate real-time data logging, knowledge inferencing, simulation and tutoring. This enabled modeling phenomena such as the incorporation of instruction, acquisition of attentional focus, and increased parallelism among perception, cognition, and action. In the present effort, we will explore training opportunities offered by such a learning simulation. In particular, we will experimentally research the use of agent-based instruction for tutoring real-time complex task performance. This effort will provide a prototype of how to use a leaning simulation to improve training of real-time dynamic tasks, a critical need for a broad range of 21st Century Navy aviator jobs.