ABSTRACT
ABSTRACT Past research suggests that the ability of air traffic controllers to determine whether two target aircraft will pass too close to each other is mediated by the convergence angle between them: wider convergence angles are associated with impaired conflict detection (especially with high traffic loads), due to the greater required amount of visual scanning. We attempted to generalize these results in more ecologically valid displays (a) with broader altitude distributions than those used in previous studies; and (b) where controllers couldn’t be certain that a conflict existed on every trial. Fourteen air traffic controllers performed a standard conflict detection task, during which both convergence angle and traffic load were systematically manipulated. Though we observed large effects of traffic load on both accuracy and response times, convergence angle did not affect performance, even at high loads. Thus convergence angle may play only a limited role in mediating controllers’ conflict detection performance in many actual air traffic displays. Keywords: air traffic control; visual scanning; convergence angle; conflict detection INTRODUCTION Air traffic controllers provide for the safe, orderly, and expeditious movement of traffic in the national airspace system. The provision of this service is highly dependent on the ability of the controller to extrapolate the trajectories of aircraft from their current positions to a future point and then to spatially determine whether or not their distance at the closest point of approach violates the minimum separation requirement. If it does, a conflict exists. A controller’s conflict detection ability is in turn affected by a variety of factors that include (but are not limited to) traffic load, aircraft type, airspace geography, sector size, and convergence angle. Here we focus primarily on the role which the convergence angle between two aircraft may or may not play in mediating conflict detection. The possible importance of convergence angle in this context is suggested by intuition, by previous research in other domains, and by anecdotal aspects of salient failures of conflict detection in the real world. Intuitively, it seems likely that a greater convergence angle between two aircraft will lead to greater difficulty in determining the likelihood of a conflict between them. This intuition is based primarily on the necessary step of scanning between the two aircraft in order to detect a conflict (e.g.. Bisseret, 1981): wider convergence angles will increase the time and distance of this scanning, and may often result in the encountering of more intervening distractor aircraft. As a result, it seems natural to expect that high convergence angles may lead to conflict detection impairments in terms of both response time and accuracy. Such predictions are also supported by empirical studies of the role of convergence angles in other contexts that needn’t directly involve conflict detection. For example, wider convergence angles cause subjects to take longer when simply extrapolating trajectories (e.g. Smith, Ellis, & Lee, 1984).
