The Effects of REM Sleep Deprivation on the Level of ...

SLEEP DEPRIVATION

The Effects of REM Sleep Deprivation on the Level of
Sleepiness/Alertness

Keith Nykamp, Leon Rosenthal, Mike Folkerts, Timothy Roehrs, Peter Guido, and Thomas Roth
Henry Ford Hospital, Sleep Disorders and Research Center, Detroit, Mich

Study Objectives: The purpose of this study was to evaluate the effects of acute REM deprivation on daytime sleepi-
ness/alertness, as measured by the MSLT.
Participants: Twenty-six healthy, normal volunteers (14 males and 12 females) participated in this study. Participating
subjects were in good physical and psychological health and were asymptomatic as to sleep/wake complaints.
Interventions: Subjects spent 5 nights and 5 days in the laboratory. The first night and day were utilized for screening
purposes. The remaining stay in the laboratory consisted of a baseline night and day, 2 deprivation nights and days, and
a recovery night and day. Each night, a nocturnal polysomnogram was employed to monitor subjects’ sleep. Each day,
subjects underwent an MSLT to evaluate their sleepiness/alertness. Subjects were randomized into REM-deprivation (RD)
and yoked-control (YC) groups. On deprivation nights, RD subjects were awakened each time they entered stage REM
sleep, and the YC subjects were awakened concomitantly with the RD subjects, assuming they were not in stage REM
sleep.
Results: The REM-deprived subjects did not demonstrate any changes in MSLT scores across experimental days. In
contrast, the YC subjects documented significantly lower MSLT scores on deprivation days due to decreased total sleep
time.
Conclusion: The REM-deprivation procedure antagonized the effects of sleep loss on daytime sleepiness, resulting in
increased alertness for RD subjects compared to YC subjects. The mechanism by which REM deprivation exerts its alert-
ing effects is unknown and will require future research.
Key words: REM sleep; REM deprivation; sleepiness; alertness; CNS excitability

WITH THE ADVENT of the multiple sleep latency test In addition to daytime sleepiness, acute sleep depriva-
(MSLT),1 researchers have objectively characterized the tion has been documented as having profound effects on
effects of total sleep loss,2 sleep restriction,3 and sleep dis- the length of recovery sleep. Using an enforced 24-hour
ruption4 on daytime sleepiness. These studies have con- recovery period, a 72% recovery of lost sleep has been
firmed earlier subjective reports that sleep loss and sleep demonstrated.7 Changes in the composition of the sleep
disruption increase daytime sleepiness. Indeed, acute architecture have also been demonstrated following sleep-
sleep-deprivation studies have documented a linear rela- deprivation protocols. Dement et al demonstrated that a
tionship between nocturnal sleep and daytime sleepiness.5,6 reduction in total sleep time over several nights resulted in
an increased percentage of slow-wave sleep (SWS) during
Accepted for publication June, 1998 recovery sleep.8 Borbély et al demonstrated a significant
increase in SWS intensity on 2 contiguous recovery nights
Address correspondence and requests for reprints to Leon Rosenthal, MD, following 40 hours of total sleep deprivation.9
Henry Ford Hospital, Sleep Disorders and Research Center, 2799 W. Grand Furthermore, extended sleep in the morning,10 and late-
Blvd., CFP 3, Detroit, MI 48202. afternoon naps,11,12 have been shown to significantly

SLEEP, Vol. 21, No. 6, 1998 609 Effects of REM deprivation on sleepiness—Nykamp et al

decrease the amount of SWS during nocturnal sleep when Table 1. The screening sleep characteristics of the REM-deprived and
compared to baseline levels. Daytime naps have also been- yoked control groups
shown to facilitate the reversal of daytime sleepiness fol-
lowing acute sleep deprivation. Lumley et al positively REM- Yoked Significance
correlated the length of daytime naps and the amount of deprived control
SWS present during the naps with mean MSLT scores after group t=1.21, p ns
sleep loss.13 group t=1.13, p ns
94±3 t=1.18, p ns
In contrast to SWS sleep, REM sleep appears to be pri- Sleep efficiency (%) 93±3 t=1.43, p ns
marily determined by circadian variations rather than puta- Stage 1 (%) 9±3 12±9 t=.411, p ns
tive homeostatic responses to acute total sleep loss.14-21 Stage 2 (%) 55±11
During 24-hour nap studies, the presence of REM sleep Stage 3/4 (%) 18±5 56±7
was more dependent on the time of day, whereas SWS was Stage REM (%) 18±3
more dependent on the degree of sleep restriction and the ± standard deviation 15±8
length of prior wakefulness.16,17 Indeed, acute total sleep
deprivation only minimally disrupts the 90-minute cycle of 17±5
REM sleep.3,9,20 However, Bennington and Heller suggest-
ed that REM sleep propensity accumulates during NREM METHODS
sleep, and should only occur following NREM sleep.22
They described a linear function directly relating the Subjects
amount of NREM sleep to the amount of REM sleep. Alas,
in normal, healthy subjects, REM sleep has been demon- Twenty-six healthy, normal volunteers (14 males and
strated to occur every 90 minutes throughout nocturnal 12 females) participated in the study. Subjects responded
sleep only after sufficient NREM sleep.23,24 In addition, to advertisements presented in local newspapers, and were
the amount of REM sleep also appears to be a function of screened over the telephone in order to establish that they
previous REM deprivation. Selective REM-sleep-depriva- had regular sleep/wake schedules (ie, an average of 6.5-8.5
tion studies, while minimizing the loss of NREM sleep, hours of sleep time per night without habitual napping), no
have demonstrated a substantial increased propensity for recent drug use, and no medical or psychiatric conditions.
REM sleep, as manifested by a significant rebound in REM In addition, subjects were required to pass a physical exam-
sleep on recovery nights.8 In rats, REM deprivation results ination, a psychological evaluation, and a urine toxicology
in most of the same “sleep deprivation effects”25 as total analysis prior to entry into the study. Participants in the
sleep deprivation and NREM sleep deprivation, including study signed a written consent form explaining the require-
increased energy expenditure and immunosuppression. ments of participation and the monetary compensation they
Both human and animal REM-deprivation studies indicate would be receiving for their participation.
that REM sleep is a vital homeostatic process. REM depri-
vation has also been shown to result in a general enhance- Procedures
ment of activity, increased sexual activity, and decreased
electroconvulsive-seizure threshold in healthy subjects.26-28 Participating subjects were in good health as deter-
mined by a screening physical examination, and were
The available data point to REM sleep as a necessary asymptomatic as to sleep/wake complaints. An 8-hour
component of sleep/wake homeostasis. The animal litera- nocturnal polysomnogram (PSG) was employed for the
ture has described the vital role of REM and NREM sleep purposes of screening subjects for sleep disturbances. A
in the survival of the organism. The human literature has four-nap MSLT was administered the following day to
demonstrated the importance of total sleep time and the assess subjects’ sleepiness/alertness. Subjects had no evi-
consequences of sleep restriction on the level of daytime dence of a sleep disorder on screening PSG and had a mean
sleepiness. Studies have also suggested the particularly MSLT score ³8 minutes (mean = 13.0±3 minutes).
important role of SWS in the alleviation of sleepiness. In Subjects with REM onset on the MSLT were excluded
contrast, the role of REM sleep during acute sleep-depriva- from the study. Subjects spent the next 4 nights and days
tion studies remains unclear. In fact, no study has system- in the laboratory. They were randomized to the REM-
atically evaluated the effect of REM sleep manipulations sleep-deprivation group (RD; n=13) or to the yoked-control
on the level of daytime sleepiness/alertness. The present group (YC; n=13). If possible, participation in the study
study was designed to evaluate the effects of acute REM- was coincident for each pair of experimental and yoked
sleep deprivation on the level of daytime sleepiness/alert- subjects.
ness, as measured by the MSLT.

SLEEP, Vol. 21, No. 6, 1998 610 Effects of REM deprivation on sleepiness—Nykamp et al

Table 2.—The sleep characteristics of the REM deprived (RD) and yoked control (YC) groups

Baseline Deprivation 1 Deprivation 2 Recovery

RD YC RD YC RD YC RD YC

TST (min) 443±19 456±15 308±86 308±39 282±42 309±40 454±25 451±19

Latency to 1 (min) 18±11 13±10 30±23 31±16 31±30 28±20 15±18 21±18

Stage 1% 10±5 12±8 21±9 19±9 17±6 16±8 8±2 12±8

Stage 2% 54±8 54±7 52±12 46±9 53±9 50±8 48±9 50±12

Stage 3/4% 16±9 14±8 22±9 15±10 24±10 16±9 21±10 19±13

Stage REM% 19±1 20±1 4±1†* 19±1 5±1†* 18±1 24±2 23±2

Latency to REM (min) 76±32 86±33 75±38 87±39 62±18 91±42 57±12 90±54

± standard deviation †vs Baseline & Recovery, p<.01
*vs YC, p<.01

Subjects were required to refrain from consuming alco- Kales,30 with an interrater reliability ³ 90 percent. Subjects
hol and caffeine for 5 hours prior to arrival on the screen- were asked to complete the Positive Affect/Negative Affect
ing night and throughout the study. On the screening night Scale (PANAS) and the Profile of Mood States (POMS)
and each of the 4 experimental nights, electrodes were three times each day (10:30, 14:30, and 16:30) throughout
placed on subjects for unipolar monitoring of the central the study. PSG variables, MSLT scores, and mood vari-
and occipital electroencephalograms (EEG), electrooculo- ables were analyzed using the Multivariate General Linear
grams (EOG), submental electormyogram (EMG), and an Hypothesis testing utilities of Systat, version 5.2 for
electrocardiogram (EKG). Tibialis anterior EMG to moni- Macintosh (Evanston, Ill: Systat, Inc., 1992). The between-
tor periodic leg movements, and a thermistor placed at the group variable was group (RD or YC), and the repeated
nose and mouth to monitor respiration, were used to screen measure was the 4 consecutive experimental nights/days.
for sleep apnea. On each of the study days, an MSLT was Probabilities reported were corrected for multiple compar-
administered to determine subjects’ level of sleepiness/ isons using the Greenhouse-Geiser method. Tukey’s post
alertness. The standard MSLT protocol was utilized for hoc comparisons were performed when appropriate.
each MSLT.29 On the screening day, subjects were asked at
10:00, 12:00, 14:00, and 16:00 hours, to lie down, close RESULTS
their eyes, and attempt to fall asleep. On experimental days
(1-4), five naps were administered (10:00, 12:00, 14:00, Both of the groups were comparable, with seven males
16:00, and 18:00 hours). Naps were terminated following and six females each, and had comparable ages of 21±3
20 minutes of wakefulness or 15 minutes after the first years. There were no group differences in sleep character-
epoch of sleep. Between naps, subjects were monitored to istics on screening night (see Table 1).
ensure that they remained awake.
Sleep Characteristics on the Experimental Nights
Time in bed (TIB) on nights 1 and 4 (baseline and
recovery) was 8 hours (23:30 to 07:30 hours), and all sub- The sleep characteristics were submitted to a two-fac-
jects were allowed to sleep undisturbed. On nights 2 and 3 tor ANOVA, with one between-group factor (RD and YC)
(deprivation nights), subjects in the RD group were exper- and one repeated-measures factor (experimental nights).
imentally awakened following one epoch of unequivocal RD subjects were awakened 9±2 times during the first
stage-REM sleep. They were kept awake for 15 minutes deprivation night, which was comparable to the number of
before being allowed to return to sleep. Each time RD sub- experimental awakenings for YC subjects (8±2). The num-
jects were awakened, the YC subjects were also awakened ber of awakenings during the second deprivation night was
if they were not in stage REM. If the YC subjects were in significantly higher (p<.01) for both RD (11±3) and YC
REM sleep, the awakening was delayed until they entered (10±3) subjects. Experimental awakenings were compara-
NREM sleep. TIB on nights 2 and 3 was extended to 9 ble between RD and YC subjects on the second deprivation
hours and 15 minutes (22:15 to 07:30 hours) as compensa- night.
tion for the loss in total sleep time due to the experimental
awakenings. Nocturnal PSG recordings were scored in 30- Total sleep time (TST) did not reveal a significant main
second epochs according to the rules of Rechtschaffen and effect of group or group-by-night interaction (see Table 2).
However, the experimental awakenings during the depriva-

SLEEP, Vol. 21, No. 6, 1998 611 Effects of REM deprivation on sleepiness—Nykamp et al

tion nights resulted in a significant decrease in TST Figure 1.—The MSLT scores (mean ± SEM) of the REM-deprived
(F[3,72]=121.5, p<.01). On recovery night, TST was com- (RD) and yoked control (YC) groups
parable to baseline. The latency to stage 1-NREM sleep
(minutes) did not reveal a significant main effect of group, scores on deprivation days 2 (11.6±4 minutes) and 3
but did reveal a significant main effect of night (11.0±4 minutes) when compared to baseline (13.7±3 min-
(F[3,72]=6.41, p<.01). Post hoc comparisons revealed the utes) and recovery (15.0±4 minutes) days was documented.
latency to stage 1-NREM on baseline and recovery nights Critically, a significant group-by-night interaction was doc-
to be comparable but significantly shorter (p<.05) than umented (F=6.29, p<.01) for MSLT scores. RD subjects
those on experimental nights. The 2 experimental nights demonstrated comparable MSLT scores across days (base-
were comparable. A group-by-night interaction was not line=12.9±3 minutes; deprivation day 1=13.3±3 minutes;
documented for latency to stage 1-NREM. deprivation day 2=12.2±3 minutes; recovery=14.9±2.7
minutes). However, YC subjects documented lower MSLT
Percent stage 1 NREM was significantly higher scores on deprivation days 1 (9.8±4 minutes) and 2 (9.8±4
(F[3,72]=21.12, p<.01) on the 2 deprivation nights than minutes) when compared to baseline (14.5±3 minutes) and
either baseline or recovery. Deprivation night 1 was high- recovery (15.1±2.4 minutes) days. Furthermore, when
er than deprivation night 2 (p<.05). There was no effect of compared to RD subjects, YC subjects had significantly
group or group-by-night interaction. Stage 2 results show lower (p<.05) MSLT scores on deprivation days 1 and 2
a significant night effect (F[3,72)=4.33, p<.01). Percent (see Fig. 1).
stage 2 was higher on baseline than any other night. This
difference was significant (p<.05) in comparison to depri- A nap-by-nap analysis was performed for the daytime
vation night 1 and recovery, but not deprivation night 2. MSLTs. There were no group differences in sleep latency
There was again no effect of group or group-by-night inter- patterns across the days. However, a significant main
action. SWS revealed a significant main effect of night effect of nap (F[4,96]=27.7, p<.01) was documented. The
(F[3,72]=6.20, p<.01), but no group effect or interaction. latency to sleep for the nap at 10:00 hours (7.9±3.5 min-
Post hoc comparison revealed that SWS percent was lower utes) was significantly shorter (p<.05) than the latency to
on baseline than any other experimental night. sleep on naps at 12:00 hours (12.0±4.2 minutes) and 14:00
hours (12.5±3.7 minutes). The nap at 16:00 hours
Importantly, a significant main effect of group (14.6±4.1 minutes) documented a significantly longer
(F[1,24]=21.78, p<.01), a significant main effect of night (p<.05) latency to sleep than the naps at 12:00 and 14:00
(F[3,72)=37.66, p<.01), and a significant group-by-night hours. Finally, the nap at 18:00 hours (17.1±2.4 minutes)
interaction (F[3,72]=19.00, p<.01) was documented for documented a significantly longer (p<.05) latency to sleep
REM percent. The percentage of REM sleep for RD sub- than the nap at 16:00 hours. A group-by-nap interaction
jects was significantly reduced (p<.01) from baseline levels was not documented.
on both deprivation nights (see Table 2). Recovery REM
sleep was significantly higher than on both deprivation An analysis of the number of sleep-onset REM periods
nights, but only slightly higher (p<.08) than baseline. The 2 (SOREMPs) present during MSLTs was carried out, and
deprivation nights showed comparable REM sleep. In con- there were not any significant effects of group or night, or
trast to the RD subjects, the percentage of REM sleep for group-by-night interaction (see Table 3).
YC subjects was similar on all experimental nights.
Tukey’s post hocs revealed significant (p<.01) group dif-
ferences in the percentage of REM sleep on nights 2 and 3
(see Table 2). Latency to REM was also affected by the
REM-deprivation procedure. A significant main effect of
group (F[1,24]=4.23, p<.05) was documented for latency to
REM. The RD group exhibited shorter latencies to REM
sleep when compared to YC subjects. The RD group had
an overall REM latency of 56.8±11.4 minutes, while the
YC group’s REM latency was 89.6±53.8 minutes. There
was not a significant main effect of night or a significant
group-by-night interaction for REM latency.

Daytime Assessment

MSLT scores did not reveal a main effect of group, but
did demonstrate a main effect of night (F[3,72]=15.91,
p<.01). A significant (p<.05) decrease in mean MSLT

SLEEP, Vol. 21, No. 6, 1998 612 Effects of REM deprivation on sleepiness—Nykamp et al

Table 3.—The number of subjects with multiple sleep-onset REM periods Previous studies have suggested that selective REM
during their MSLT deprivation results in enhanced CNS excitability. Cohen
and Dement first described the effects of REM deprivation
Baseline Deprivation Deprivation Recovery as being “a generalized increase in neural excitability.”28
day 1 day 2 They demonstrated that after rats were deprived of REM
RD group 0 0 sleep, their threshold to electroconvulsive shock signifi-
YC group 0 1 2 1 cantly decreased. In a subsequent paper, the same group
1 1 described increases in sexual activity as a result of REM
deprivation.31 Furthermore, this general increase in CNS
Mood Assessment drive following REM deprivation has also been manifested
by increased motivational behaviors,27 increased aggres-
The PANAS scores were subjected to a three-factor sion,32 and increased preference for novel stimuli.33 In
ANOVA (group ´ experimental day ´ time). There was not contrast to these studies, subjects restricted in total sleep
a significant main effect of group (RD=21±7; YC=24±9) time and/or deprived of NREM sleep demonstrated
on the positive affect (PA). The PA scores did reveal a sig- decreased motivation and performance retardation.34-36
nificant main effect of experimental day (F[3,51]=10.56, More importantly, sleep restriction and sleep fragmentation
p<.01, G-G<.01). The average PA score for baseline (24±8) have resulted in increased daytime sleepiness, as manifest-
was comparable to the average score for recovery (26±9). ed by the MSLT.3-5 In spite of comparable numbers of
The average PA scores for deprivation days 1 (21±7) and 2 awakenings and sleep loss on experimental nights, the
(20±8) were also comparable. However, PA scores on REM-deprived group in the present study maintained its
deprivation days were significantly lower (p<.05) than the level of sleepiness/alertness, while the yoked control group
PA scores on baseline and recovery. A significant time-of- experienced increased daytime sleepiness.
day effect (F[2,34]=3.70, p<.05, G-G<.05) was also docu-
mented for PA scores. The average PA score at 14:30 Bennington and Heller (1994) theorized that the prima-
(22.3±7.6) was significantly lower (p<.05) than the average ry function of REM sleep is to antagonize the restorative
PA score given at 10:30 (22.8±7.7). The average PA score effects of NREM sleep. Moreover, NREM sleep is charac-
given at 16:30 (22.5±8.0) was intermediate to the other two terized by synchronized EEG waves and an inert CNS,
scores. No significant interactions for PA scores were doc- whereas REM sleep is characterized by desynchronized
umented. The negative affect (NA) did not reveal any sig- EEG waves and a highly reactive CNS.37 Thus, if the func-
nificant group effects, day-of-experiment effects, or time- tions of REM sleep and NREM sleep appear to be antago-
of-day effects. The POMS questionnaire was administered nistic, and the physiology of REM sleep and NREM sleep
concomitantly with the PANAS questionnaire; however, are paradoxical, it is plausible to hypothesize that the
none of the POMS categories revealed any significant effects of REM deprivation should be antipodal to the
group, day, or time effects. effects of NREM deprivation. In fact, the results of the pre-
sent study largely confirm this hypothesis.
DISCUSSION
A previous study by Glovinsky et al documented a
The present study evaluated daytime sleepiness/alert- shortening of MSLT latencies for subjects awakened during
ness in subjects selectively deprived of REM sleep. stage 2 sleep and for subjects awakened during REM
Subjects with awakenings yoked to the REM-deprivation sleep.38 The authors concluded that a “reduced sleep time
subjects were also evaluated for daytime sleepiness/alert- instead of altered sleep composition was responsible for the
ness in order to control for the effects of the nocturnal observed daytime decrements.” In the Glovinsky et al
awakenings. The yoked control group experienced study, subjects were only required to remain awake for 3
increased sleepiness due to the increased number of awak- minutes following REM awakenings, resulting in 12 and 18
enings and overall sleep loss on deprivation nights. The experimental awakenings on deprivation nights 1 and 2
REM-deprived group did not exhibit any changes in MSLT respectively. As a consequence of these awakenings, TST
scores throughout the study. It should be noted that both was reduced to 331 minutes, but REM sleep was only
the REM-deprived and the yoked control groups docu- reduced to 9% of TST. In contrast, subjects in the present
mented an increase in nocturnal latency to stage 1 NREM study were required to remain awake for 15 minutes fol-
sleep on both deprivation nights. This is not, however, a lowing REM awakenings, resulting in only 9 and 11 awak-
function of the experimental awakenings, but rather an arti- enings on deprivation nights 1 and 2 respectively.
fact of the earlier bedtimes for subjects on deprivation Interestingly, this resulted in a mean TST of 295 minutes
nights to account for reduced total sleep time. and a significant reduction in REM sleep to 4.5 percent of
TST. It is conceivable that the effect of sleep fragmentation
and sleep loss in the study by Glovinsky et al inundated the

SLEEP, Vol. 21, No. 6, 1998 613 Effects of REM deprivation on sleepiness—Nykamp et al

alerting effect of the REM deprivation, whereas in the pre- phase on nap sleep. Electroencephalog Clin Neurophysiol 1986;
sent study, the CNS excitability caused by significant REM
deprivation compensated for the loss of sleep which result- 64:224-227.
ed in no changes in sleepiness/alertness from baseline lev-
els. 18. Serge D, Beersma DGM, Borbely AA. Timing of human sleep:

In summary, the results of this study emphasize the del- recovery process gated by a circadian pacemaker. Am J Physiology
icate balance between NREM and REM sleep homeostasis.
While an increase in CNS excitability is suggested as 246:R161-R178.
explanatory for the compensation of alertness seen in the
RD group, the mechanism by which REM deprivation 19. Beersma DGM, Dijk DJ, Blok CGH, Everhardus I. REM sleep
exerts its effects on CNS excitability has not been exten-
sively explored. An elucidation of these mechanisms may deprivation during 5 hours leads to an immediate REM sleep rebound
pave the way for a greater understanding of the function of
REM sleep. and to suppression of non-REM sleep intensity. Electroencephalog Clin

REFERENCES Neurophysiol 1990; 76:114-122.

1. Carskadon MA, Dement WC. Sleep tendency: an objective mea- 20. Borbély AA. A two process model of sleep regulation. Human
sure of sleep loss. Sleep Research 1977;6:200.
2. Carskadon MA, Dement WC. Effects of total sleep loss on sleep Neurobiology 1982; 1:195-204.
tendency. Perceptual and Motor Skills 1979;48:495-506.
3. Carskadon MA, Dement WC. Cumulative effects of sleep restric- 21. Czeisler CA, Zimmerman JC, Ronda JM, Moore-Ede MC,
tion on daytime sleepiness. Psychophysiology 1981;18:107-113.
4. Stepanski E, Lamphere J, Badia P, Zorick F, Roth T. Sleep frag- Weitzman ED. Timing of REM sleep is coupled to the circadian
mentation and daytime sleepiness. Sleep 1984;7:18-26.
5. Carskadon MA, Dement WC. Daytime sleepiness: quantification rhythms of body temperature in man. Sleep 1980; 2:329-346.
of a behavioral state. Neuroscience and Biobehavioural Reviews 1987;
11:307-317. 22. Benington JH, Heller HC. Does the function of REM sleep concern
6. Rosenthal L, Roehrs T, Rosen A, Roth T. Level of sleepiness and
total sleep time following various time in bed conditions. Sleep 1993; non-REM sleep or waking? Progress in Neurobiology 1994; 44:433-
16:226-232.
7. Rosenthal L, Merlotti L, Roehrs T, Roth T. Enforced 24-hour 449.
recovery following sleep deprivation. Sleep 1991;14:448-453.
8. Dement W, Greenberg S, Klein R. The effect of partial REM sleep 23. Dement WC, Kleitman N. Cyclic variations in EEG during sleep
deprivation and delayed recovery. J Psychiat Res 1966;4:141-152.
9. Borbely AA, Baumann F, Brandeis D, Strauch I, Lehmann D. Sleep and their relation to eye movements, body motility, and dreaming.
deprivation: Effect on sleep stages and EEG power density in man.
Electroencephalog Clin Neurophysiol 1981;51:483-493. Electroendephalogr Clin Neurophysiol 1957;9:673-690.
10. Feinberg I, Fein G, Floyd TC. Computer detected patterns of elec-
troencephalographic delta activity during and after extended sleep. 24. Dement WC. The effect of dream deprivation. Science
Science 1982;215:1131-1133.
11. Feinberg I, March JD, Floyd TC, Jimison R, Bosson-Demitrack L, 1960;131:1705-1707.
Katz PH. Homeostatic changes during post nap sleep maintain baseline
levels of delta EEG. Electroencepholog Clin Neurophysiol 1985; 25. Rechtschaffen A, Bergmann BM, Everson CA, Kushida CA,
61:134-137.
12. Karacan I, Williams RL, Finley WW, Hursch CJ. The effects of Gilliland MA. Sleep deprivation in the rat: X. Integration and dis-
naps on nocturnal sleep: influence on the need for stage-1 REM and
stage 4 sleep. Biol Psychiatry 1970;2:391-399. cussion of the findings. Sleep 1989;12:68-87.
13. Lumley M, Roehrs T, Zorick F, Lamphere J, Roth T. The alerting
effects of naps in sleep deprived subjects. Psychophysiology 1986; 26. Whitehead WE, Robinson TM, Wincor MZ, Rechtschaffen A. The
23:403-408.
14. Webb W, Agnew H, Sternthal H. Sleep during the early morning. accumulation of REM sleep need during sleep and wakefulness.
Psychon Sci 1966;6:277-278.
15. Webb W, Agnew H. Sleep cycling within twenty-four hour Comm in Behav Biol 1969;4:195-201.
episodes. J Experimental Psychol 1967;74:158- 160.
16. Carskadon MA, Dement WC. Distribution of REM sleep on a 90 27. Vogel GW. A review of REM sleep deprivation. Arch Gen
minute sleep-wake schedule. Sleep 1980;2:309-317.
17. Dinges DF. Differential effects of prior wakefulness and circadian Psychiatry 1975;32:749-761.

28. Cohen HB, Dement WC. Sleep: changes in threshold to electro-

convulsive shock in rats after deprivation of “paradoxical” phase.

Science 1965;150:1318-1319.

29. Carskadon MA, Dement WC, Mitler M, Roth T, Westbrook P,

Keenan S. Guidlines for the multiple sleep latency test (MSLT): a stan-

dard measure of sleepiness. Sleep 1986;9:519-524.

30. Rechtschaffen A, Kales A, eds. A manual of standardized termi-

nology, techniques and scoring system for sleep stages of human sub-

jects. Washington, DC: U.S. Government Printing Office, 1968.

31. Dement W, Henry P, Cohen H, Ferguson J. Studies on the effect of

REM deprivation in humans and in animals. In: Sleep and altered states

of consciousness. The Williams and Wilkins Company, Baltimore,

1967;456-469.

32. Hicks RA, Moore JD, Hayes C, Phillips N, Hawkins J. REM sleep

deprivation increases aggressiveness in male rats. Physiol Behav

1979;22:1097-1100.

33. Moore JD, Hayes C, Hicks RA. REM sleep deprivation increases

preference for novelty in rats. Physiol Behav 1979;23:975-976.

34. Williams HL, Lubin A, and Goodnow JJ. Impaired performance

with acute sleep loss. Psychol Monogr 1959;73:1-26.

35. Kollar CJ, Pasnau RO, Rubin RT, Naitoh P, Slater GG, and Kales A.

Psychological, psychophysiological and biochemical correlates of pro-

longed sleep deptrivation. Am J Psychiatry 1969;126:488-497.

36. Dinges DF, Kribbs NB. Performing while sleepy: effects of exper-

imentally-induced sleepiness. In: Monk T, ed. Sleep, sleepiness, and

performance. John Wiley and Sons, 1991.

37. Dujardin KA, Guerrien A, Leconte P. Sleep, brain activation and

cognition. Physiol behav 1990;47:1271-1278.

38. Glovinsky PB, Spielman AJ, Carroll P, Weinstein L, Ellman SJ.

Sleepiness and REM sleep recurrence: The effects of stage 2 and REM

sleep awakenings. Psychophysiology 1990;27:552-559

SLEEP, Vol. 21, No. 6, 1998 614 Effects of REM deprivation on sleepiness—Nykamp et al