The 757/767 has two pilot seats. In airline operations, the captain sits in the left seat and the first officer in the right seat. ``Seat'' thus becomes a metonym for role in the cockpit. Captains have a wider scope of authority and responsibility than first officers. They also normally have more flight experience than first officers. We are interested in how these factors may affect attitudes toward automation. Airlines operate from bases where crew trips (possibly consisting of many flight segments over several days) begin and end. We speculated that differing kinds of flight operations flown from different bases, for example principally international or domestic operations, might affect crew attitudes toward automation.
In recent years, the backgrounds of pilots coming to the airlines has changed. The proportion of pilots with military training has decreased. Some in the industry have expressed concern about this trend. We wondered if there are any differences in attitudes toward automation among pilots who received flight training from the various branches of the military and those who had no military flight training.
Finally, we are interested in the effect of experience on attitudes toward automation. In the world of flying, hours of experience is the primary measure of expertise. We measured three types of experience: total flying hours, hours in type (that is, in the 757/767), and hours in other automated airplanes (airplanes equipped with a Flight Management Computer System). Two additional measures of experience were created in later analysis procedures. First, time in other FMS equipped aircraft was added to time in type to produce a measure of total experience in FMS equipped aircraft. Second, the ratio of FMS equipped time to total time was computed for each subject. Imagine two pilots who each has 3000 hours of experience in automated airplanes. One has 23,000 hours of total time and the other has 5,000 hours of total time. We expect that some attitudes toward automation might depend as much on the proportion of one's career spent in automated airplanes as on the total number of hours in automated airplanes.
We felt that automated hours would be the most important of these measures of experience in determining attitudes toward automation. The cultural convention in the pilot community is to specify flight experience in hours, and all responding subjects did so. One problem with the survey was that it did not include the Airbus A300 as an explicit prompt on the questionnaire. Twenty nine of the responding pilots used the 'other' entry to indicate that they had experience in this FMS equipped airplane.
The amount of calendar time it takes to acquire a given number of hours may be different for different pilots. Pilots with enough seniority to bid and hold a steady schedule may be able to fly about 80 hours per month. Pilots with less seniority may not alway be able to fly as many hours as they desire. In case attitudes toward automation are formed by time spent thinking about the system as well as time spent actually operating it, we also measured calendar time since completion of training in the airplane. We chose to partially replicate Wiener's 1989 study [9]. We retained the format of Wiener's instrument, presenting each probe as an assertion over scaled responses (a modified Likert scale): ``strongly agree'', ``agree'', ``neutral'', ``disagree'', ``strongly disagree''. We chose the 15 probes that were most closely related to autoflight from the original 36 probes in Wiener's study.
Appendix 2 provides a list of the probes and the abbreviations for them that will be used throughout this report.
In some cases, we made slight changes to the wording of the questions to accommodate our plans to use the attitude survey with pilots flying airplanes other than the 757/767. For example, Wiener's probe, ``The B-757 automation works great in today's ATC environment.'' was reworded to read ``The automation in my current aircraft works great in today's ATC environment.''
We added one new probe to the survey. This item is ``I always consult the flight mode annunciator to determine which mode the autopilot/flight director is in.'' Over the past decade, the automation in airplanes has become more complex. The issue of ``mode awareness'' that is, the crew's understanding of what the automation is doing, has been cited as a causal factor in a number of accidents [2,4]. Because of this, airlines and airframe manufacturers have stressed the importance of consulting the flight mode annunciator. However, observations of pilots and conversations with them indicate that pilots may not use the flight mode annunciators in the way intended. We hoped that this probe would help us understand the relationship of consulting the flight mode annunciators to other attitudes toward automation. We also recognized from the outset that pilots might be tempted to respond to this probe by reporting what they know they are supposed to do rather than what they actually do. We have no way of knowing how many pilots dissembled on this probe. We do know that several of them indicated that they did not always consult the FMA, but wrote in the margins of the survey that they knew they should.
The survey also included a set of prompts for judgments of the similarities among the names of autoflight modes that appear on the flight mode annunciator in the airplane. The results of that part of the study will be presented in a later report.