Adaptive driving simulation-based training: framework, status, and needs

Motor vehicle crashes are the major cause of deaths in the United States. Conventional driving training approaches including classroom education, vehicle owner manuals, training videos, on-the-road skill training, and non-adaptive driving simulation-based training have limited effect on crash and traffic violation reduction. The objectives of this research were to: (1) develop a framework for adaptive driving simulation-based training (ADST); (2) review the recent advances in ADST; (3) provide recommendations to improve adaptive training in this area; and (4) identify the domains that can benefit from ADST. A systematic literature search was conducted using Compendex, Web of Science, and Google Scholar databases to find relevant research published since 2000. Results indicated that the ADST framework consists of three components including performance measures, adaptive logic, and adaptive variables. Adaptations can occur on the training content and/or feedback and prior (i.e., parametering) or during the training session (i.e., dynamic adaptation). Based on the developed framework and literature review, a set of guidelines was provided to improve ADST along with some specific domains that can benefit from this approach. Finally, recommendations were provided to evaluate ADST effectiveness and transfer of knowledge in future.