Additionally, it is possible that the circadian pattern of skill-based
errors observed in this study reflected changes in the nature of the
tasks over the 24 h. Compared to workers on day shifts, workers on
night shifts may have received less supervision, may have had access to
more limited resources, been faced with a less favorable work
environment (e.g., in terms of lighting levels or access to colleagues), or
been assigned more simple skill-based or self-paced tasks. While these
possibilities cannot be completely discounted, field observations and
discussions with maintenance personnel suggest that task complexity, at
least, did not vary systematically across the hours of the day. As aviation
is a 24 h industry and aircraft maintenance occurs around the clock, it
is likely that maintainers are not at liberty to schedule specific tasks and
must deal with problems when they arise in order to keep planes
flying. Finally, information on the duration of time the error-makers
had been awake was not collected. Therefore, we cannot distinguish
endogenous circadian effects from sleep/wake-related factors involved
in the observed 24 h error patterns. These two factors independently,
and in interaction, affect the profile of cognitive performance across the
24 h