A Comparison of the Micro-Adaptive and Hybrid Approaches to Adaptive Training

Abstract

Recently, there has been a push to investigate state-of-the-art approaches in computer-based training (CBT) that have the potential to be as effective as one-to-one tutoring. One promising method is Adaptive Training (AT). However, there are a number of different AT approaches as well as numerous variables that can be adapted and scant research to guide training system designers. In our study, we compared four different AT approaches within the context of a military task. While there were no significant differences in performance scores during training, our learning gain results support the notion that a Hybrid Approach to AT is more effective than a Micro Approach during the most difficult training scenarios and regardless of the variables that are adapted. It is possible that this result is due to a reduction in germane cognitive load when the starting difficulty matches a learner’s spatial ability. Implications for the design of future AT systems are discussed.