People often think new drivers slowly get safer as they gain additional experience. Our research shows it is more complicated. The Phase Transition Theory suggests that new drivers’ risk of crashing can change suddenly at different times for different drivers, rather than everyone gradually improving with practice. For example, researchers have observed that changes in strategy, that is, how someone approaches a task, can lead to big jumps in performance. It may be that on average, certain strategies are better than others but at the same time specific strategies might not work equally well for everyone.

By following each driver’s progress closely across many different domains instead of only looking at group averages, we can better understand when and why these change points happen for different drivers. This knowledge can help us create better laws, training, and technologies so we can help new drivers learn to drive safely from the beginning.

Peer-reviewed research about the Phase Transition Model

• Learning to drive: A reconceptualization

• A dynamical systems perspective on driver behavior