Search Results

Large-scale simulations are making satisfactory progress in all areas except human performance. Graphics have produced human models that appear realistic (face validity), but do not always behave realistically. With a wealth of data about humans and human performance available, what keeps better data and models from being integrated into large-scale simulations? Do new requirements for verification, validation, and accreditation help or hinder the integration of the human component into large-scale simulation? Can those researching and modeling physical and cognitive issues work together to integrate their technologies? Can human performance models supplement or replace human-in-the-loop simulation, thereby reducing the cost and time required to develop a new system? The issues and technologies involved in developing and integrating human performance models are discussed.

The increased need for better human performance models results from a realization that comprehensive simulations cannot use human-in-the-loop for all players, as was originally thought. With system designers and decision makers demanding support from large-scale simulation models, the human performance component of computer models has repeatedly been found to be deficient. Each of the five communities involved with human performance modeling communicates with adjacent communities in a pair-wise manner, causing the customer to have little awareness of what is technically possible and the scientific community having little awareness of the needs of the customer.