Lisa C. Thomas

Human factors issues in implementation of advanced aviation technologies: a case of false alerts and cockpit displays of traffic information

This paper provides a framework for identifying and evaluating the human performance implications of new avionics technology such as the cockpit display of traffic information (CDTI). Several important human factors issues in developing and implementing the CDTI are discussed in terms of their effects on the two primary flight tasks associated with the CDTI: conflict detection and conflict resolution. One issue is the development of alerting algorithms, which are defined by how uncertainty, false alarms and look-ahead times are dealt with. Another issue concerns understanding how time pressure, display dimensionality and conflict geometry affect pilot preferences for certain avoidance maneouvres over others. A third issue is the proposed shift in responsibility of flight safety away from air traffic control to the pilot after CDTIs make it possible for pilots to perform tasks related to self-separation. These issues and the factors affecting them are classified into several categories of dependent and independent variables and placed into a framework illustrating their effects on false alarm rates and human tolerance for false alarms. A review of the literature reveals gaps in understanding of the impact of these issues under various operational situations and new directions for research are proposed.

Immersion and Battlefield Visualization: Frame of Reference Effects on Navigation Tasks and Cognitive Tunneling

Army officers viewed a battle scenario in one of two computer-based display conditions. The Tethered condition was a 3-D exocentric display and the Immersed condition was a 3-D egocentric display, which allowed 360° panning, coupled with a small 2-D contour map of the entire battle area embedded in the top center of the screen. Tasks included providing verbal reports of new enemy units or changes to existing units, responding to a series of diagnostic questions, and giving confidence ratings for their selected answers. Results showed that Immersed officers were less accurate on questions that required panning as well as questions requiring a count of enemy units than Tethered condition officers. However, confidence ratings for both tasks did not differ between display conditions. These findings indicate that Immersed officers were “cognitively tunneled” into the initial forward field of view, relying too heavily on information in this view and not adequately panning the environment.

Effects of battlefield display frames of reference on navigation tasks, spatial judgements, and change detection

This paper describes an experiment which illustrates the cause of ‘cognitive tunnelling’ as it affects information gathering and perception-based task performance in computer-generated terrain displays of varying frames of reference. Cognitive tunnelling refers to the effect where observers focus attention on information from specific areas of a display to the exclusion of information presented outside these areas. Previous research suggests that cognitive tunnelling is induced by more immersive or egocentric visual displays. Results from our preceding study suggested that an immersed split-screen display induces cognitive tunnelling and results in poorer information extraction and situation awareness than an exocentric display of the same information. The current study determined that failure of the observers to integrate information across the two views of the immersed display led to the cognitive tunnelling effect. Cognitive tunnelling was also affected by primacy of information initially presented within the larger egocentric view in the immersed display.

Display Dimensionality and Conflict Geometry Effects on Maneuver Preferences for Resolving In-Flight Conflicts

Objective: Two experiments explored the effects of display dimensionality, conflict geometry, and time pressure on pilot maneuvering preferences for resolving en route conflicts. Background: With the presence of a cockpit display of traffic information (CDTI) that provides graphical airspace information, pilots can use a variety of conflict resolution maneuvers in response to how they perceive the conflict. Inconsistent preference findings from previous research on conflict resolution using CDTIs may be attributable to inherent ambiguities in 3-D perspective displays and/or a limited range of conflict geometries. Methods: Pilots resolved predicted conflicts using CDTIs with three levels of display dimensionality; the first had two 2-D orthogonal views, the second depicted the airspace in two alternating 3-D perspective views, and the third had a pilot-controlled swiveling viewpoint. Results: Pilots demonstrated the same preferences that have been observed in previous research for vertical over lateral maneuvers in low workload and climbs over descents for level-flight conflicts. With increasing workload the two 3-D perspective displays, but not the 2-D displays, resulted in an increased preference for lateral over vertical maneuvers. Increased time pressure resulted in increased vertical maneuvers, an effect again limited to the two 3-D perspective displays. Conclusion: Resolution preferences were more affected by workload and time pressure when the 3-D perspective displays were used, as compared with the 2-D displays, although overall preferences were milder than in previous studies. Application: Investigating maneuver preferences using the strategic flight planning paradigm employed in this study may be the key to better ensure pilot acceptance of computer-generated resolution maneuvers.

Redundancy Gains in Communication Tasks: A Comparison of Auditory, Visual, and Redundant Auditory-Visual Information Presentation on NextGen Flight Decks

The redundant presentation of information in more than one sensory channel has traditionally been assumed to benefit performance. However, a recent meta-analysis suggests that redundancy gains may depend on task type and a number of moderator variables. The present study examined the effectiveness of visual, auditory, and redundant auditory-visual information presentation under high and low workload in the context of a mid-fidelity NextGen flight simulation with experienced airline pilots as participants. Overall, for two types of communication tasks – data link and ATIS – faster responses were observed for redundant displays, compared to vision and audition alone. No significant benefit of redundancy was found for accuracy due to a ceiling effect and workload did not mediate redundancy effects. The findings from this research add to the knowledge base in multimodal and redundant information processing and can inform modality choices in the design of displays for complex, data-rich domains.

Display Dimensionality and Conflict Geometry Effects on Maneuver Preferences for Resolving in-Flight Conflicts

With the presence of a cockpit display of traffic information (CDTI) that provides graphical airspace information, pilots can use a variety of conflict resolution maneuvers in response to how they perceive the conflict. Inconsistent preference findings from previous research on conflict resolution using CDTIs may be due to inherent ambiguities in 3-D displays and/or a limited range of conflict geometries. This paper describes a study that investigates conflict resolution maneuver preferences using three displays with different frames of reference and a wide range of conflict geometries. Results indicate that 3-D displays with interactive viewpoints reduced spatial ambiguities. The interactive 3-D displays produced a preference for vertical maneuvers over lateral similar to a 2-D coplanar display; however this preference was reversed under increased workload conditions for both 3-D displays. Pilots in all three display conditions showed a preference to maneuver vertically away from intruders, though this was eliminated (or reversed) as workload increased. The 2-D coplanar display induced a preference to laterally turn away from approaching intruders, which overwhelmed a trained “turn right” preference.

Multimodal Information Presentation in Support of NextGen Operations

ABSTRACT Objective: This study examined the effectiveness of visual, auditory, tactile, and redundant auditory-visual information presentation in the context of a medium-fidelity ‘Next Generation Air Transportation System’ (NextGen) flight simulation. Background: Data overload, especially in the visual channel, and associated breakdowns in monitoring represent a major challenge in aviation. These problems are expected to worsen with NextGen, which will require pilots to manage increased amounts of data and adopt new responsibilities. The introduction of multimodal interfaces (interfaces that distribute information across multiple sensory channels) has been proposed as a means to offload the overburdened visual channel and thus address data overload. Method: Experienced commercial airline pilots completed 2 scenarios using a medium-fidelity flight simulator. For each scenario, NextGen tasks and events were presented either using technology that is currently available (visual and auditory displays) or technology proposed as part of NextGen design concepts (i.e., tactile and redundant displays). Performance was measured based on response time and accuracy. Results: Faster responses were observed for redundant displays, compared to either vision or audition alone. No significant benefit of redundancy was found for accuracy and workload did not mediate redundancy effects. For traffic events, there were faster response times with tactile displays, but higher response accuracy with auditory displays. Conclusion: The findings from this research add to the knowledge base in multimodal information processing and can inform the design of displays for NextGen operations.