Newly designed equipment and procedures often inadvertantly faciliate human error. Techniques for identifying error facilitations in design tend to be either ineffective or very expenswe. For example, one of the most effective ways to test new designs is to hire human operators to carry out tasks using prototyped equipment, and then observe their performance in a wide range of operating conditions. In our domain, air traffic control, such tests typically require hiring highly paid expert controllers as subjects, often for extended periods. The limited amount of testing that results from high cost stifles innovation and compromises safety.