Pre-attentive Features in Natural Augmented Reality Visualizations

The movement towards cyberphysical systems and Industry 4.0 promises to imbue each and every stage of production with a myriad of sensors. The open question is how people are to comprehend and interact with data originating from industrial machinery. We propose a metaphor that compares machines with natural beings that appeal to people by representing machine states with patterns occurring in nature. Our approach uses augmented reality (AR) to represent machine states as trees of different shapes and colors (BioAR). We performed a study on pre-attentive processing of visual features in AR to determine if our BioAR metaphor conveys fast changes unambiguously and accurately. Our results indicate that the visual features in our BioAR metaphor are processed pre-attentively. In contrast to previous research, for the BioAR metaphor, variations in form induced less errors than variations in hue in the target detection task.