Loss of control events often involve erroneous pilot mental models of the aircraft state and autoflight system modes in the period leading up to the event. These mental models include their immediate situation awareness of the aircraft state, but also often include their knowledge of the autoflight system. Both aspects of these have been modeled computationally in fast-time simulations. The situation awareness of aircraft state is represented as a model-based observer (e.g. Kalman filter). The knowledge of autoflight modes is represented as a finite state machine where each state represents a different control behavior, and transitions between modes are explicitly represented to describe the conditions in which they can be commanded by the pilot and/or commanded automatically. This paper will describe the results of simulations in which the mental model is run dynamically through flight events potentially leading to loss of control.