Objective:

Human factors design depends on accurate models of human task performance. Without such models, it would be impossible to optimize a control panel layout or choose one keyboard configuration over another without building prototypes and testing them with large numbers of users. This research subtask is developing user-friendly extensions of the Apex human-system modeling framework. Apex improves on ACT-R, which added overlapping activities to the Goals, Operators, Methods, and Selection-rules (GOMS) method published by Card, Moran, and Newell in 1983.

Applications:

Modeling of human task learning and performance for man-machine interface design, air traffic control, or launch and range operations.

NASA Benefit:

Better control panel design can enable faster data entry (with less operator fatigue), better situation monitoring, and reduced risk of human error. This research subtask will greatly reduce the time and expertise needed for such design, with accurate generic models of human task performance.

Keywords:

cognitive models, interface design, HCI, task learning, human usability performance testing

Prior Technology:

ACT-R is an improvement over GOMS for modeling human task performance, allowing for overlapping activities. Development of task models is still time-consuming and requires special expertise.

Parent Task:

Launch and Range Operations Decision Models.

Related Web Pages:

CogTool.

Contacts:

Bonnie E. John (PI), Carnegie Mellon University.