Perhaps the most important conclusion to
emerge from the findings reviewed in this paper is
that there are consistent temporal peaks in accident
risk, or ‘black times’, associated with both time of
day and time on task. When traffic density is controlled
for, accident risk shows a major peak at ca
03:OO and a secondary one at ca 15:O0. This time of
day effect in accident risk is extremely similar to that
found in industrial performance measures, suggesting
that it may be due to the individuals’ circadian
rhythms rather than to the very different driving
conditions pertaining at night. Sleep propensity shows
a marked circadian rhythm which, during daylight
hours, shows a very similar trend to that in performance
efficiency on a range of tasks. However,
detailed analysis suggests that these circadian rhythms
are insufficient to account for the variation in accident
risk over the 24-hour day. In particular, there would
appear to be three ‘residual’ peaks in risk at ca 02:00,
14:OO and 21:OO that cannot be accounted for in
terms of the known circadian rhythm in sleep
propensity.