Ecological Interface Design (EID) is a recent philosophy for designing the visual displays of human-machine interfaces. An EID interface displays the higher-order relations and properties of a work domain so that adaptive operator problem solving is better supported for both normal and abnormal system conditions. Previous empirical studies of EID have assumed that the raw data required to derive and communicate the higher-order information would be available and reliable. The present research empirically evaluates the impact of having incomplete data on the effectiveness of an EID interface, compared to a more traditional piping and instrumentation interface. The research also addresses recent criticism directed at previous empirical studies of EID is also addressed. Results suggest that diagnostic performance using an ecological interface is compromised only when the interface is supported by a minimal set of instrumentation. However, an ecological interface supported by maximal instrumentation, on average, leads to the best diagnostic performance.
