Sleepiness
is not always perceived before falling
asleep
in
healthy, sleep-deprived subjects
Herrmann US, Hess CW, Guggisberg AG
Roth C, Gugger M, Mathis J
Department of Neurology,
Inselspital, Bern University
Hospital
and University of Bern,
Switzerland.
OBJECTIVE: To test whether subjects
spontaneously signal sleepiness before falling
asleep under monotonous conditions.
METHODS: Twenty-eight healthy students were
deprived of sleep for one night and then
underwent a "maintenance-of-wakefulness test"
(MWT) consisting of four 40-min trials. They
were told to give a signal as soon as they felt
sleepy and to try to stay awake as long as
possible. In a first series of tests, the
subjects were given no reward (nr); in a second
series, monetary rewards (wr) were given both
for an accurate perception of sleepiness and for
staying awake longer.
RESULTS: Seventeen of the 28 subjects
(60.7%) did not signal sleepiness before a sleep
fragment occurred in at least one of the four
MWT trials. Women were more reliably aware of
sleepiness than men in the nr trials (p=.02),
while the men's performance improved in the wr
trials (p<.02), becoming equivalent to the
women's performance.
CONCLUSIONS: Our results cast doubt on the
general assumption that one cannot fall asleep
without feeling sleepy first. If similar results
can be obtained in monotonous driving or working
situations, this will imply that accidents
caused by sleepiness or by falling asleep cannot
necessarily be attributed to an individual's
negligence.
1. Introduction
Work accidents and motor-vehicle accidents
(MVAs) are often caused by sleepiness or by
falling asleep. Drivers involved in MVAs of this
type usually deny having fallen asleep at the
wheel or having felt sleepy before the crash.
These subjective, retrospective judgments are
generally considered false and attributed to the
faulty recollection of sleepiness, rather than
to the true absence of sleepiness or of feeling
sleepy, J.Maiiy authors suggest'thatit' is
nearly impossible to fall asleep without being
aware of sleepiness beforehand, because certain
physiological and cognitiv&indicators of
sleepiness, such as straining, drooping eyelids,
yawning, and wandering thoughts invariably
precede the onset of sleep. Others, however,
argue that sleepiness impairs driving
performance in part by lessening the driver's
insight into performance deficits, even though
the warning signs of sleepiness are, in fact,
perceived. A driver might also underestimate the
grave risk of sleepiness at the wheel because of
previous episodes of nodding off for a few
seconds without incident. Driving while drowsy
and transiently falling asleep at the wheel are
known to happen far more often than
sleepiness-rela crashes.
Driving-simulator studies ha repeatedly
shown that drivers who fall asleep at the wheel
and drift out of their driving lane do, in fact,
perceive their momentary level of sleepiness
fairly well, although they may underestimate the
likelihood that they will fall asleep. Other
studies, however, have found only a weak or
moderately strong correlation between subjective
sleepiness ("sleep complaints") and objective
measures such as microsleep. We suggest that
these conflicting results are due to differences
in the experimental task conditions. In most
studies, the researchers asked drivers about
their subjective sleepiness before or after
testing or else at regular intervals during the
test. This differs from the actual situation
before an MVA, in which the driver can become
aware of his or her own sleepiness only by
spontaneous introspection. Repetitive,
externally timed questioning can itself provide
a sleepy person with continuous feedback
information about his or her alertness and
thereby increase the accuracy of subjective
judgments of sleepiness. Repetitive questioning
also directly makes the subject more alert.
Moreover, many studies that compared subjective
sleepiness with objective measures were
performed in driving simulators. We suggest that
interaction with the environment, by providing
continuous feedback on performance errors, helps
the subject become aware that sleep is
imminent.
The main goal of this study is to determine
whether healthy subjects can fall asleep without
spontaneously signaling sleepiness (SPS) before
a sleep fragment lasting longer than 3 s when
the trials are carried out under very monotonous
conditions, with no interaction with the
environment or feedback information on
performance allowed. The subjects' later
recollection of sleepiness played no role in
this study; subjects were precisely instructed
how to signal their imminent subjective
sleepiness just oqe before the start of each
trial of the maintenance-of-wakefulness test
(MWT). They could not interact with the
experimenters during the trial, and their
perception of sleepiness was thus based purely
on spontaneous introspection.
These MWT conditions are admittedly more
passive than an average driving situation, and
it is thus unclear to what extent MWT results
can be extrapolated to driving situations.
Subjects might fall asleep more easily under the
MWT conditions, but in real-life situations the
risk of falling asleep might be higher in
general because of the longer exposure times.
MWT is widely available in clinical sleep
laboratories and is generally accepted as the
gold standard for measuring
maintenance-of-wakefulness. It thus seems to be
a reasonable compromise solution to the problem
of assessing both the subjects' ability to stay
awake and their subjective perception of
sleepiness (SPS).
As the risk of an accident in a real-life
driving or working situation motivates people to
behave responsibly, it would seem desirable to
have a penalty for bad performance in the
experimental setting as well. Because punishing
experimental volunteers is not feasible, we
introduced, in a second series, a monetary
reward for goal achievement to improve the
subjects' motivation.
We recruited men and women in roughly equal
numbers to test whether women might be better
than men at recognizing their sleepiness, which,
if so, might explain their lower risk of
accidents at work or behind the wheel.
4. Discussion
The main finding of this study is that
sleep-deprived healthy young subjects often did
not spontaneously signal sleepiness (SPS) before
falling asleep under the monotonous MWT
conditions. This late, rather than appropriate,
SPS was seen in 37.5% of all trials when defined
as the absence of signaling before any sleep
fragment lasting longer than 3 s. The risk of
late SPS was highest in subjects who suffered
the greatest daytime sleepiness from sleep
deprivation, which at first glance seems
paradoxical. The finding makes sense, however,
if we assume that spontaneous introspection and
reporting of sleepiness are themselves impaired
by progressing sleepiness. Hardly any late SPS
was found with longer sleep latencies of 20 min
or more. This corroborates the commonly used
criterion of a mean sleep latency of 15-20 min
or longer in the MWT for the ability to
drive.
Our results differ from those of earlier
studies, in which subjective sleepiness was
found to be closely correlated with objective
measures of sleep pressure or impaired
performance. We argue that these differences can
be explained largely by differences in the
experimental methods used.
First, many previous studies. employed the
sleep-onset criteria of Rechtschaffen and Kales,
in which episodes of microsleep shorter than 15
s were disregarded. If we had used this
criterion for overt sleep (OS) and disregarded
all sleep fragments shorter than 15 s, we would
have observed late, rather than appropriate, SPS
in only 10.7% of all trials (12.5% of nr trials
and 9.4% of wr trials). It is clear, however,
that in demanding tasks requiring a high degree
of wakefulness, such as surveillance tasks or
driving, sleep fragments as short as 3 s or less
could have disastrous consequences, and
microsleep is known to be much more common than
overt sleep during daylight hours.
Second, unlike in previous studies, the
subjects in this study were not repeatedly
questioned during the trials. Instead, they were
instructed only at the beginning of the MWT
trials to signal imminent SPS. Thus, no abrupt
changes of alertness were induced by external
interaction, and sleepiness was perceived purely
by spontaneous introspection.
Third, in contrast to the situation in
experiments on driving simulators or other
performance tasks, our subjects had no feedback
information on their performance.
We hypothesize that our subjects might have
assessed their own sleepiness more accurately
than they did in our SPS protocol if they had
been allowed to interact continuously with
examiners and if they had received feedback
information about their performance.
Kaplan et al. also found that predictions
about falling asleep were particularly poor for
the very first sleep fragment, and they also
suggested that their subjects misjudged the
relevance of their sleepiness indicators.
In our study, sleepiness was not signaled at
all during the full 40 min in only 7.1% of
trials: according to the KSS obtained after each
trial, sleepiness was undetected in only 2%.
Thus, our subjects signaled sleepiness in almost
all MWT trials,, but they often did so a few
seconds up to a few minutes after a short sleep
fragment of greater than 3 s duration. Thus, our
results do not contradict the trivial fact that
one is sleepy afthr a night of sleep
deprivation: Rather, they suggest that the
timing of the subjective judgment of sleepiness
is not precise in the dynamic situation of a
rapid fluctuating alertness level on a
second-by-second basis.
To our knowledge, this is the first
systematic study of the awareness of
spontaneously perceived sleepiness under MWT
conditions. It challenges the doctrine that one
cannot fall asleep without prior perception of
sleepiness, at least under monotonous MWT
conditions without external interaction and
without feedback about performance errors.
These results cannot automatically be
generalized to everyday situations, to other age
groups, or to sleepy persons suffering from a
clinical disease. If, however, future studies
reveal that similar results are generally
applicable in a wide variety of driving or
working situations, there could be major
consequences both with respect to the forensic
aspects of sleepiness-induced motor-vehicle or
work accidents and with respect to the
instructions that physicians should give to
their sleepy patients. Until now, from a legal
point of view, sleepiness-induced work or
motor-vehicle accidents have been attributed to
reckless behavior and punished accordingly,
because it was assumed that these drivers
continued to drive despite having perceived
sleepiness. Our results, however, imply that
this approach might be questionable under
certain conditions such as absence of external
interaction and feedback information of
performance errors.
The main limitation of this exploratory
study is the small number of subjects, all of
whom were young. Further studies should be
performed in older subjects, in persons who have
undergone less severe sleep deprivation, and in
sleepy persons with clinical diseases. The
results in any of these groups might differ from
the ones obtained here. Clinical experience
suggests that persons who are chronically sleepy
because of illness often underestimate their
sleepiness.
Moreover, our results only permit limited
conclusions as to why the subjects did not
signal their sleepiness before the first sleep
fragment, even though video recordings of the
face regularly showed obvious indicators of
sleepiness, such as yawning, straining, and
drooping eyelids. Our interpretation (that the
subjects did not, in fact, perceive sleepiness
before falling asleep) was supported by the
subjects' statements at the end of the test day.
Only one subject explained her failure to
appropriately signal sleepiness by saying that
she had waited to become even sleepier in order
to be more certain about the reason for
signaling. A learning effect was not observed,
as the frequency of late SPS was independent of
the trial number over the course of the test
day.
The sex difference found here requires
cautious interpretation in view of the small
number of subjects. An earlier study also showed
that women perceive sleepiness better than men.
Theoretically, this might explain why young men
are more commonly involved than young women in
sleepiness-induced accidents at work and at the
wheel.
Further studies under different conditions,
including surveillance tasks or driving
simulators, are needed to elucidate the role of
subjectively perceived sleepiness in the
causation of accidents. We also propose
including and assessment of SPS-and, even more
importantly, of micros leep-in the routine MWT,
in order to gain a better understanding of the
impact of late SPS in patients at risk of work
or motor-vehicle accidents, despite a normal MWT
latency.
-Giganti F, Hayes
MJ Cioni G, Salzarulo P Yawning frequency
and distribution in preterm and near term
infants assessed throughout 24-h recordings
Infant Behav & Development
2007;30(4):641-647
