The Firebreak is Gone: Why Faster Flight Training Content Needs a Firmer Foundation

In a moment of frustration, I recently made a post on LinkedIn about the accelerating rate at which people are churning out new flight training resources and platforms. I noted these were largely using AI, but I wanted to clarify my intent. My concern is the approach, and AI is only one part of it. Many of the people who are building these new platforms are experienced pilots and instructors. There are people with thousands of hours flying and thousands as instructors. They are working on the very real problem that instructors are stretched to breaking point and bear the load of facilitator, instructional designer and teacher, all while remaining a skilled pilot at the same time. The aim deserves plenty of respect, but it is not without risk.

The risk is that products offered quickly, even in good faith, are only working from what we have right now. They are just another repackaging of the same problems. AI is the accelerant on a fire that has been smouldering since World War 2. Content is now cheap, information in abundance, and ground school material can be generated nearly instantly, with quizzes, visual aids, summaries and scenarios. Sure enough, the same could be done by hand, but the time and effort it took acted as a firebreak. It’s gone now. Healy et al. (2014) describe training like a balancing act where methods that speed up early acquisition can reduce how long a skill lasts, and that even a durable skill does not always transfer to novel situations. Landman et al. (2018) also found that training in unpredictable and variable scenarios improved transfer to unexpected events. What this means is that neatly packaged and predictable content might be pulling effective training in the wrong direction.

The other risk I see is that any approach which takes more of the social component out of learning and promises faster and cheaper, is still only offering the same content without instructional design or a firm foundational philosophy underneath. Humans learn through well proven, if unglamorous, theories from educational psychology. Aviation has applied them to human factors and non-technical skills. It is yet to link them as directly to how we teach pilots. The Ecological Dynamics Model for Aviation (EDMA™) proposes a model to make that link explicit. It describes learning as the relationship between the person and their environment, rather than inside the learner’s head alone (Davids, 2012). Humans come to know the world by perceiving it and acting in it. A pilot learns to read energy states, attitudes and vectors by flying and feeling how the aircraft responds. Davids (2012) also argues that practice must represent the conditions where the skill is likely to be used, and that practice which routinely isolates skills from their context falls short. A library of reading and quizzes can support understanding, and discrete skills are worth knowing, but in isolation and without context, learning may fall short.

Instructors do far more than deliver information. They watch, correct, debrief, calibrate confidence and model how a pilot thinks. When a product promises to take the load off the instructor, it may inadvertently remove the real teaching and learning. An instructor who teaches poorly only affects their students. While not acceptable, it is somewhat contained or slowed. A product sold widely online carries any gaps or errors to every student who uses it, at a fast rate. In a recent online program, I saw material teaching that power controls airspeed on approach with no mention that pitch is the only airspeed control once the engine stops. In a domain where the cost of an error can be death, educational oversight is part of safety.

AI fits into the flight training space and can offer some exciting things. The value may lie less in regeneration of old material. Maybe it is AI’s ability to tailor learning to the individual, or to lift the administrative and compliance burden on instructors? I use AI and it has wonderful benefits. Before any of us build with it, however, I suggest we ask ourselves to check in. Here are some starter questions:

1.      What will the learner perceive and do, and how close is that to the real aircraft environment?

2.      How does it build variability and surprise into practice?

3.      How is retention planned for after the module ends?

4.      Where does the instructor sit in the learning, and what does the product leave to them?

5.      Who checked the material against educational evidence and aviation resources, and how?

6.      How does the student identify and fix errors?

A product that can answer these will add value to the flight training landscape. A product that cannot is adding height to a structure whose foundation is still unfinished. That foundation is what EDMA was built to address. The paper is free to read and is available under Creative Commons on our website.

Davids, K. (2012). Learning design for nonlinear dynamical movement systems. The Open Sports Sciences Journal, 5(1), 9-16. https://doi.org/10.2174/1875399X01205010009.

Healy, A.F., Kole, J.A., & Bourne, L.E., Jr. (2014). Training principles to advance expertise. Frontiers in Psychology, 5, Article 131. https://doi.org/10.3389/fpsyg.2014.00131.

Landman, A., van Oorschot, P., van Paassen, M. M., Groen, E. L., Bronkhorst, A. W., & Mulder, M. (2018). Training pilots for unexpected events: A simulator study on the advantage of unpredictable and variable scenarios. Human Factors, 60(6), 793–805. https://doi.org/10.1177/0018720818779928.