“…Most importantly, it is the first study to be able to explain these performance breakdowns based on converging behavioral, eye-tracking, and mental model data. Our findings call for more conceptual and exploratory approaches to training (e.g., Casner, 2003;Chappell, Crowther, Mitchell, & Govindaraj, 1997;Mumaw, Boorman, & Griffin, 2001;Mumaw, Boorman, Griffin, Moodi, & Xu, 2000;Sarter & Woods, 2000) and improved feedback design, which could take the form of multimodal interfaces (see Nikolic, Orr, & Sarter, 2004;Sarter, 2000;Sklar & Sarter, 1999;D. Javaux, personal communication, 2004) and visualizations of automation intent and aircraft behavior (see Boorman & Mumaw, 2004), to avoid future mode errors and automation surprises.…”