Bridging Cognitive Modeling and Pilot Training: A Qn-Actr Model for Complex Flight Tasks

Authors

Rongbing Xu, University of Waterloo
Shi Cao, University of Waterloo
Elizabeth Irving, University of Waterloo
Suzanne Kearns, University of Waterloo
Ewa Niechwiej-Szwedo, University of Waterloo

Document Type

Article

Publication Date

2025

Abstract

In ab initio pilot training, students must learn substantial skills about how to fly anaircraft, but there has been a lack of research that can clearly model these skills.Cognitive architecture models use production rules to model skills, which can beapplied to the case of pilot training to facilitate skill diagnosis, assessment, andfeedback. In this study, researchers who have both flight training experience andcognitive modelling experience developed cognitive architecture models for takeoffand landing tasks using the Queueing Network-Adaptive Control of ThoughtRational (QN-ACTR) method. The models generated flight performance in XPlaneflight simulation similar to human pilot in-aircraft data from the same tasksrecorded in the same type of aircraft (Cessna 172). This presentation details theproduction rules, assumptions, validation results, and lessons learned. Thesefindings provide the foundation for future work to further develop the models forother flight tasks and pilots with different levels of experience.

Comments

Presented at the 23rd International Symposium on Aviation Psychology, May 27-30, 2025, Hosted by Oregon State University

Repository Citation

Xu, R., Cao, S., Irving, E., Kearns, S., & Niechwiej-Szwedo, E. (2025). Bridging Cognitive Modeling and Pilot Training: A Qn-Actr Model for Complex Flight Tasks. Proceedings of the 23rd International Symposium on Aviation Psychology, 198-203.
https://corescholar.libraries.wright.edu/isap_2025/34