Beyond the Wizard of Oz: Using Imperfect Machine Learning to Examine the Impact of Reliability of Augmented Reality Cues on Visual Search Performance

Searching for an object in their visual field, even when the person knows exactly what the object looks like, is a complex task. This is time-consuming and can be error-prone. We present findings of automation bias when imperfect cues are presented using real-time machine learning, presented in the Augmented Reality head-mounted display using the YOLOv5 machine learning model. We validated that Wizard-of-Oz studies produce results that are equivalent to using a real-time machine-learning system for visual search aided by cues.