| Summary: | Cognitive projection enables the operator of a supervisory control system, such as air traffic control, to
use predicted future behavior of the system to make decisions about if and how to control the system.
New procedures and technologies being implemented in the air traffic control system innately affect the
information used for projection and the type of projection required from the controller. Because cognitive
projection is not well-understood, launching these projection-impacting technologies and procedures
could result in the reluctance of the air traffic controllers to accept these advancements or limit the system
performance.
A Projection Process Model and a Projection Error Concept were proposed to describe the controller’s
projection process and the contextual system influences on the projection process. The two primary
influences on the projection process were the information/display system and task-based projection
requirements. A mismatch between the information/display system states and the task-based projection
requirements was described through a cognitive transform concept. The projection process itself is
composed of the state mental model and the time into the future over which the projection is made.
Hypotheses based on the assumptions of the Projection Process Model and Projection Error Concept were
probed through an experiment using an ATC task paradigm. Results were consistent with the proposed
models. They suggested that the controllers were able to incorporate higher-level dynamics into the state
mental models used for projection and that the quality of the state mental model used was marginally
influenced by the error tolerance required in the task.
The application of the Projection Process Model and Projection Error Concept was then illustrated
through the analysis of the impact on projection from two ATC domain examples of technology and
procedure implementation. The Constant Descent Approach Procedure in the TRACON impacted the
intent, projection timespan, and abstractions used in the mental model of the controllers. The Oceanic
ATC surveillance, communication and workstation improvements resulted in an impact on the states to be
projected, intent, projection timespan, and human/automation projection responsibility. Suggestions for
improved transition for the projection process were then provided based on the analysis.
|
|---|