The focus of this project is to develop training procedures for knowledge and skill that will survive primary task or background changes and, thereby, produce flexible, as well as efficient and durable, performance in emergency situations both in civilian life and in the military. The proposed experiments examine performance in dynamic and changing task environments. A complex set of concurrent tasks will be created analogous to those encountered by digitally proficient operators who have available fully computerized control and communication systems. Two experiments will be conducted within this paradigm. One of them will focus on a concurrent duration estimation task, specifically the different forgetting functions for declarative and procedural information in this task. In the second experiment, a proposed memory constriction principle will be tested, according to which the time span from which knowledge can most easily be retrieved and used in a given context shrinks as stress level increases. The broader impacts of the proposed experiments are in the effective utilization of training skills for emergency response and reaction that persist over time. This is especially important for everyone involved in homeland security since training is likely to be sporadic.