Using automated state space planning for effective management of visual information and learner’s attention in virtual reality

Educational immersive virtual reality is often tasked with minimising distractions for learners and maintaining or signalling their focus to the right areas. Managing location, density and relevancy of visual information in the virtual environment pertain to this. Essentially this problem could be defined as the need of management of cognitive load from the visual information. To aid in the automated handling of this problem, this study investigates the use of automated state-space planning to model the current “state” of the virtual environment, and determine from a given pool of steps or “actions”, a sequence that prioritise minimising cognitive load from visual information through planning the location and density of objects. This study also investigates modelling the state of what a learner has been informed of and applied. This enables planning to determine when to have the learner relate concepts to existing knowledge for deeper knowledge; planning their generative learning. These states are planned in conjunction with the virtual environment states. The planning is also responsive to identified changes in the learner’s deviated attention, or performance with the task. Together it has the potential to minimise the cognitive load from being taught intrinsic information, and minimising extraneous information from the virtual environment. What was produced currently does not yield many results beyond the method of planning helping the virtual reality applications manage where information appears, but it at least also established a framework for future testing, and improvements to the used methods. This paper provides in more detail, the background for this topic in immersive virtual reality, its significance, the methods used and an evaluation of the method and how further investigations will be continued.