Aviation psychology. II: Assessing workload and selecting pilots.

The principal impact on the crew of improved technology on the flight deck has been a reduction in the number of tasks that the crew must perform but an increase in the amount of information that must be monitored. Under such circumstances it is difficult to determine the extent of the pilot's workload, yet its assessment is crucial in deciding how many people are required to operate a given aircraft. In a military interceptor the benefits of having an additional person on board may be offset by the extra airframe weight and performance penalties that his presence entails. Airline economics dictate the presence of only the minimum number of flight crew commensurate with safe operation, and in a modern airliner the question becomes whether to have two or three people on the flight deck. These matters are overlaid with so many politicoindustrial considerations that an objective means of assessing workload is clearly desirable. Indeed the "two v three" crew issue was so strongly debated in the United States that a president's task force on crew complement was set up.' The report of this committee has stimulated the Federal Aviation Administration to be satisfied that workload has been formally assessed on any new aircraft before it will be certificated. It does not, however, answer the psychologist's problem of how to carry out such an assessment. Many techniques have been suggested, but these resolve themselves into methods that require the aircrew to give some form of rating of how hard they feel themselves to be working (the most famous of these being the Cooper-Harper scale2), secondary task techniques,3 observational techniques in which the overt activity of the aircrew is analysed,4 and the measurement of physiological variables, principally heart rate.' All of these methods have severe deficiencies: subjective ratings are open to subjective bias, secondary tasks are intrusive, observation cannot indicate the cognitive load on the pilot, and heart rate is more affected by the pilot's perception of how critical the task is than by how busy it keeps him. At present, aircraft seeking certification in the United States (such as the European Airbus and the British Aerospace 146) have to use some combination of the techniques described above, but there is some encouraging basic psychological research (looking at ways of assessing how finite mental resources can be allocated) that may yield more acceptable methods of measuring workload.6-'