off. Although shift lengths vary, 10-hour shifts are not uncommon. Flight controllers average less than one break per shift (range 0.1 to 0.9) (Kelly et al., 1993).
Although the explosion of the Challenger space shuttle occurred during the daytime, the decisions made the night before the launch by mission control staff have been cited as a major factor contributing to the explosion (Mitler et al., 1988). Flight controllers are responsible for a wide range of cognitive tasks, including sustained trend analysis, monitoring of multiple voice channels, and rapid responses to emergency situations. Cognitive processing and vigilance must remain high because even small mistakes can be operationally significant (Kelly et al., 1993).
Although the aerospace industry has a history that includes several accidents and near accidents associated with fatigue, there are no regulations on how long workers employed by NASA and/or manufacturers supplying spacecraft components may work in the days prior to or during a mission. In contrast, the sleep patterns of astronauts have been monitored since the early days of the space program (Aschoff, 1965; Pittendrigh, 1967). Several passages from Apollo 13 (Lovell and Kluger, 1994) illustrate the attention paid to the sleep/wake patterns of astronauts during a mission:
After just a day or so in translunar drift, the astronauts got accustomed to the constant flickering and went about their sleep-wake, work-rest schedules as if the sun were rising and setting outside their craft just as it did outside their homes in Houston. As long as the crew maintained that schedule, NASA’s flight surgeons had learned, their circadian rhythms would remain largely un-disturbed.
Even on a routine flight, no one expected the pilots to sleep a full eight hours. The almost total lack of physical exertion in space and the almost constant output of adrenaline that accompanied the business of flying to the moon made five or six hours of sack time the most the medics could hope for. Those five or six hours, however, were absolutely essential if a crew that was flying even a nominal mission was going to make it through their day without making some serious, and perhaps disastrous, mistake. A crew that was flying a less than nominal mission would need even more rest. (Lovell and Kluger, 1994:202)
In the second row of Mission Control, the flight surgeon had been copying down the answers the men gave, and the totals had begun to alarm him. Since Monday night, the crew had been averaging about three hours of sleep apiece per day. It was 2:30 Friday morning…. (Lovell and Kluger, 1994:313)
These anecdotal reports of shortened sleep times have been confirmed by both subjective and objective studies (Dijk et al., 2001; Frost et al., 1976, 1977; Grundel et al., 1993, 1996, 1997; Monk et al., 1998; Santy et al., 1998). Despite preflight circadian adaptation measures and in-flight schedules to optimize circadian adaptation and minimize sleep loss, astronauts