Responsive Adaptive Display Anticipates Requests (RADAR) is a domain general system that learns to highlight an individual's preferred information displays, given the current context. Previous studies with human subjects in a video game environment demonstrate that RADAR is an effective cognitive aid. RADAR increases situation awareness and reduces cognitive load by anticipating and providing task relevant information. Additionally, because RADAR's fit to a user's behavior encapsulates the user's situation-driven information preferences, RADAR also excels as a descriptive and predictive assessment tool. Here, we focus RADAR as a training aid. We test the hypothesis that novices can benefit from training under a RADAR model derived from an expert's behavioral patterns. The results indicate that novices exposed to an expert's information preferences through RADAR rapidly learn to conform to the expert's preferences.
Marc T. Tomlinson, Michael Howe, Bradley C. Love