For surveillance and rescue operations, human navigation and situation awareness should be well-supported. However, proper testing of new and innovative mobile support is hard for such operations. Evaluations in field settings are expensive, difficult to control, potentially dangerous, and may sometimes be impossible to execute. Using a virtual environment may solve these disadvantages of testing in the field. But how well does the performance with mobile support in a virtual environment predict the performance with the same mobile support in the real environment? To establish validity, patterns found in situation awareness, attention and performance in the virtual environment must match with patterns found in the field. This paper describes an experiment that compares these patterns for participants who search for objects in a virtual or field (real) environment. The virtual environment seemed to be sufficient in providing the prominent cues of the field environment that people use to navigate, but navigation itself proved to be far more easy in the field so that navigation support effects differed for both environments. © 2011 International Federation for Information Processing.