Team training increasingly takes place in synthetic environments. However, team training in synthetic environments is often modeled after live team training without removing some of the disadvantages that occur in live training, such as instructor-intense performance monitoring, and the fact that all appropriate teammates have to be available. Simulated teammates are a promising alternative to human teammates, because they are always available, may be modeled after experienced training personnel, and may be more cost effective.

The Netherlands Organization for Applied Scientific Research (TNO) Human Factors and the Naval Air Warfare Center Training Systems Division are working jointly towards defining the requirements for synthetic teammates (SYNTHERs). The goals of this research effort are twofold: (1) to define the requirements for SYNTHERS and (2) to develop validated guidelines for the use of SYNTHERS in team training. In our approach to empirical validation of requirements a set of psychological experiments will be carried out, utilizing scripted humans as simulated teammates in a well-controlled simulation of a military command-and-control task using a modified version of the Dynamic Distributed Decision-Making (DDD), while taking a variety of measurements. Two experiments have been conducted so far. This paper relates the results of those experiments to an empirical validation of requirements, and provides guidelines for the design and use of SYTNHERS for team training.