The failure to find strong instructional effects of simulation is usually attributed to poor measurement and faulty research design. However, weak implementation of simulation into the classroom, rather than methodological inadequacy, is presented here as the primary reason for finding insignificant effects. The types of environments under which simulation would be expected to have its greatest impact have rarely been arranged, and thus the claims made for simulation have not been adequately tested. It is proposed that future studies should be directed toward using simulation as a vehicle for creating distinctive learning environments. Some suggestions for accomplishing this goal include: (1) establishing the phenomenological validity of the simulation, (2) creating a distinctive learning environment--using the simulation to integrate the course content, to reinforce ongoing learning, or as an advance organizer, (3) establishing the motivational context, and (4) validating simulation as an instructional process. (Author/MV)