Me. Jewett et al., Dose-response relationship between sleep duration and human psychomotor vigilance and subjective alertness, SLEEP, 22(2), 1999, pp. 171-179
Although it has been well documented that sleep is required for human perfo
rmance and alertness to recover from low levels after prolonged periods of
wakefulness, it remains unclear whether they increase in a linear or asympt
otic manner during sleep. It has been postulated that there is a relation b
etween the rate of improvement in neurobehavioral functioning and rate of d
ecline of slow-wave sleep and/or slow-wave activity (SWS/SWA) during sleep,
but this has not been verified. Thus, a cross-study comparison was conduct
ed in which dose-response curves (DRCs) were constructed for Stanford Sleep
iness Scale (SSS) and Psychomotor Vigilance Task (PVT) tests taken at 1000
hours by subjects who had been allowed to sleep 0 hours, 2 hours, 5 hours o
r 8 hours the previous night. We found that the DRCs to each PVT metric imp
roved in a saturating exponential manner, with recovery rates that were sim
ilar [time constant (T) approximate to 2.14 hours] for all the metrics. Thi
s recovery rate was slightly faster than, though not statistically signific
antly different from, the reported rate of SWS/SWA decline (T approximate t
o 2.7 hours). The DRC to the SSS improved much more slowly than psychomotor
vigilance, so that it could be fit equally well by a linear function (slop
e -0.26) or a saturating exponential function (T = 9.09 hours). We conclude
that although SWS/SWA, subjective alertness, and a wide variety of psychom
otor vigilance metrics may all change asymptotically during sleep, it remai
ns to be determined whether the underlying physiologic processes governing
their expression are different.