ABSTRACT
The appeal of simulation-based scaled worlds for performance research lies in their ability to balance realism - the ability to create simulated situations that call for complex skills and behaviors - with controlled replication - the ability to systematically manipulate factors of interest to the researcher. Scaled worlds offer a research environment that falls in between the real world, where little or no control or manipulation is possible, and classic laboratory experiments, where tasks have often been abstracted and simplified to the point where factors associated with the complexity of real-world task environments may be lost. There is a tradeoff between complexity and control in creating and using a scaled world, and this interplay creates a tension for performance measurement. Sometimes the most interesting behaviors are the most difficult to measure. Scaled worlds, however, unlike the real world or one-time exercises, offer the possibility of systematically re-creating situations that require complex behaviors, giving the researcher the opportunity to develop innovative measures and to test the reliability of these measures over multiple subjects, all confronting the same (or similar) situations.
