ABSTRACT
A major milestone in the study of human supervisory control occurred in 1948, when NormanMackworth devised the simple but very elegant clock test in order to evaluate vigilance inUKmilitary personnel (Mackworth, 1948). Participants had to observe a clock-like device and look for double-step movements (the target) among single-step movements (the non-target). Typically, Mackworth found that there was a decline in detection rate with time on task and an increase in detection time. By carrying out a series of experiments, he demonstrated that individuals could improve their vigilance performance by having shorter working periods, taking breaks, being given feedback, and in the short term, by ingesting amphetamines (Mackworth, 1950). Mackworth’s work is seminal for various reasons. It draws attention to the prob-
lem of vigilance when humans are monitoring machine-based systems. The Second World War, unlike the First World War, had been fought mainly in the air, and this had necessitated equipment being designed and implemented at great speedwithout sufficient opportunity for testing and revision. As an example, radar operators had been placed on 8-hour shifts, and Mackworth had observed that a large number of targets were being missed after a relatively short period of time. Surprising though it may seem now, knowledge about human attention span and monitoring skills, especially when checking for low frequency targets, and the application of this knowledge to the workplace, was limited. Hence, Mackworth began to lay the way by studying human interactions with technology, and in particular, the allocation of functions issues which still test us today.
