Using eye tracking to assess user behavior in virtual training

Abstract

Virtual training systems can provide flexible and effective training for interactions with increasingly complex industrial machines. However, existing approaches do not adapt to the adaptive attributes of the user. Being able to track the current state of the user enables a humanization of virtual training system, since it allows analyzing the strain and the cognitive processes of the user and reacting accordingly. In recent years, eye tracking technology has become a widespread research area in human machine interaction. This paper introduces an approach to adapt virtual training systems based on eye tracking analysis. The approach detects specific patterns from eye movements and evaluate the performance of the user based on detected patterns. If the pattern suggests that the user cannot follow the instructions or that the user is distracted, the complexity of the training system can be reduced.