On Driver Education

When we’re driving around, I often like to remind my eldest son that by the time he turns 16, in the summer of 2021, driving will probably be, or be in the process of becoming, an obsolete skill. (In the city and on the major highways, at least) Driverless cars are in the workable roadworthy prototype stage, and an array of performance support tools are being rolled out in newer cars. These tools serve to guide drivers, augment their awareness of their surroundings, and improve safety. One of the posts I currently have in draft is on the parallels between aviation and driving in terms of increasing automation and performance support. Experts tend to predict driverless cars as becoming a common reality in the early 2020s.

In the meantime, people will drive, and young people will need to learn to drive. But is it being done well? Or as well as it could. Safety equipment in cars and in the engineering of roads has reduced overall accidents and fatalities, but driving remains a relatively dangerous mode of transportation.

Young people entering into the world of driving at age 16 (and new drivers in general) need solid training to stay safe. When I turned 16 I went through the Young Drivers of Canada program. The basic program is similar to high school driver’s education, but with more emphasis on practical skills of situational awareness, threat monitoring, and collision avoidance. The training consisted of a classroom component of about one week in length, and a number of in car driving lessons. The classroom time was for teaching of theory, and consisted largely of lecture, some discussion, and watching of instructional videos. The road lessons in the car allowed practice of skills with grading and feedback by a trained instructor.

I went through this training close to 20 years ago now. This intervening period has seen a plethora of new information and sensor technologies arrive on the market. Oddly, the domain of driver education does not seem to have evolved appreciably in terms of approach. It still seems to be classroom plus supervised road lesson.

Looking at this through the eyes of an Instructional Designer, I see some gaps here in this training approach that, if filled, could lead to more effective learning, improved safety in early experiences on the road, and improved transfer and retainment of skills after the initial training is completed.

The classroom training and road practice are a good foundation, especially within a framework of defensive driving that includes threat monitoring, preventive measures, and emergency maneuvers / collision avoidance. These are necessary components of a strong foundation for an effective training system, though I’d probably advocate exploring some form of blended approach for teaching theory materials. Having some material delivered as online video or eLearning/mLearning content would allow for better flexibility of the learners, particularly adult learners. But overall, the theory and hands on supervised practice are good components of a foundation.

That said, I see two places where additional training could fill a gap and potentially improve performance.

These gaps are:

  • The need for some intermediate training to bridge the space between classroom (or online) theory and practice on actual roads. There is a wide chasm of experience between the cognitive activity of learning theory material in a classroom and the complex psychomotor skill of driving. Something in between would help to bridge this chasm and soften the transition.
  • After the course is over, there is a need for some level of electronic performance support while the young driver is out on his own. This is needed to scaffold the learner in applying the skills safely while they are still new, until such point that the skills become internalized, and automatic.

In regards to the first gap, let’s take the example of aviation. When an airline pilot is doing type training on a new aircraft, the pilot does not go directly from the classroom to flying the actual plane. First, the pilot spends time training scenarios on a simulator device. Simulation based training offers a lot of benefits. It is completely safe, but feels psychologically real when the different types of fidelity (the sense of realism of the simulation) are high. There are different dimensions of fidelity – (1) sensory perception/look and feel, (2) the process of operation of the simulated equipment versus real, and (3) the dynamics of the simulation, the relation between action and results.  It can be less expensive to operate a simulator than the actual equipment, at least once the initial investment is made to acquire the simulator. You can also control the scenario conditions to have focused training and avoid the problem practical training experiences being ad hoc, depending on the random conditions of the day.

This first gap then could be addressed through some sort of electronic driving simulator, similar to how pilots train on flight simulators. I understand that some driver training programs alresady use these, but it needs to be more universal. At one end of it could be some piece of fixed equipment similar to the flight simulator or fixed based trainer, with a mockup of a real driver’s seat, dashboard, displays/gauges, and steering wheel combined with screens to simulate out the window views, and speakers for sound effects. At the other end of it might be something like a realistic game-like driving simulation on an Xbox with a Kinect sensor and the learner’s hands as controller.

Performance support on the other hand could take the form of some sort of apparatus with cameras/sensors and a built in computer set up inside an actual car. It could collect data about speed, traffic conditions, weather, local speed limit, braking and swerving, and signalling. It could track eye movements to look out the windshield, both toward near objects and far objects, toward rear and side mirrors, and toward the blind spot. It could also monitor hand positions on the wheel (or stick, as relevant). The computer could collect data for later analysis (Syncing to a mobile app, for example).

Ideally, it would also give spoken word support cues / prompts / feedback in response to conditions and what the novice driver is doing:
“Remember to check your mirrors” “Look ahead to spot upcoming problems” “Remember to check your blind spot” “Brake!” “Try not to ride the clutch” “A little more gas” “Slow down a bit for weather/traffic conditions” “Accident ahead. Caution”)

Ideally, this form of performance support could be built into future cars as a “training / support” mode. This mode could be engaged at the push of a button on the dashboard as part of the support/automation systems that are increasingly built into modern cars. For now, though, it would have to be developed as a third party device.