SEARCH

SEARCH BY CITATION

Keywords:

  • alpha;
  • delta;
  • diurnal;
  • electroencephalogram;
  • Fast Fourier transformation;
  • homeostatic;
  • narcolepsy;
  • theta;
  • vigilance

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Narcolepsy is associated with lowered vigilance. Diurnal variation in vigilance appears altered, but the exact nature of this change is unclear. It was hypothesized that the homeostatic sleep drive is increased in narcolepsy. Decreased levels of vigilance are reflected in low frequency band power in the electroencephalogram (EEG), so these frequencies were expected to be increased in the narcolepsy group. Furthermore, it was expected that low frequency power should increase over the day. Narcoleptic patients and healthy controls participated (36 participants in total); they were not allowed to take medication or naps on the experimental day. EEG was measured at 9:00, 11:00, 13:00, 15:00, and 17:00 hours, during rest and during reaction time tasks. In the narcolepsy group, alpha power was lower at rest at all times. Delta and theta power during rest and task performance increased steadily over the day in this group, from 11:00 hours onwards. Additionally, in the narcolepsy group beta1 and beta2 power during rest appeared increased at the end of the day. The effects in the lower frequency bands strongly suggest that vigilance is low at all times. The progressive increase in low frequency power indicates that the sleep drive is enhanced. It is not clear whether this pattern reflects an extreme state of low vigilance, or a pathological brain condition. The effects in the higher frequencies suggest that narcoleptic patients may make an effort to counteract their low vigilance level.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Narcolepsy has been associated with lowered vigilance. For example, task performance is usually worse (Aguirre et al., 1985; Findley et al., 1999; Rogers and Rosenberg, 1990) and narcoleptic patients report more sleepiness (Hood and Bruck, 2002). Electroencephalogram (EEG) during wakefulness has not been studied extensively, but lower alpha power has been reported for narcoleptic patients and this has been interpreted as a vigilance decline (Honma et al., 2000). However, it is unlikely that effects are restricted to the alpha range if vigilance is indeed affected in narcolepsy. Theta power in particular is known to increase with tiredness or sleepiness (Ogilvie et al., 1991; Smit et al., in press). Therefore, increases in the low frequency bands delta and theta should also be present.

The nature of the decrease in vigilance in narcoleptic patients is not exactly clear. In order to find out what is ‘aberrant’ with vigilance in narcolepsy, it is essential to study naturally occurring changes in vigilance over the day. Moreover, narcolepsy patients may be able to shortly raise their vigilance levels to control levels (Aguirre et al., 1985), and thus multiple testing sessions during the day are preferred. In healthy individuals, there is a systematic fluctuation in vigilance throughout the day, which appears to be caused by the combined effect of circadian and homeostatic processes (Achermann and Borbély, 1994; Borbély, 1982; Carrier and Monk, 2000; Daan et al., 1984). Diurnal variations have been found in subjective alertness (Dijk et al., 1992; Johnson et al., 1992) and in task performance (Carrier and Monk, 2000; Monk et al., 1985, 1997). These measures all suggest that vigilance tends to be low in the morning and evening and higher in the late morning or afternoon. Diurnal fluctuations have also been found in the EEG. It appears that most frequency bands are associated with homeostatic processes; that is, most bands tend to show progressive declines or increases during the waking day (Aeschbach et al., 1999; Lafrance and Dumont, 2000).

Narcoleptic patients appear to have a deviant diurnal variation in vigilance. Some findings support the view that vigilance is constantly at a low level in narcoleptic patients (George et al., 1996). Other results suggest a greater postlunch dip on some measures (Pollak et al., 1992). The peak in daytime sleep propensity in narcoleptic patients appears earlier. It has been argued that this points toward impairment of the circadian process in narcolepsy (Broughton et al., 1998). However, we hypothesize that the pattern in diurnal variation may be due to an increased sleep drive. Patients, even under experimental control, are known to take naps during the day (Broughton et al., 1998; Monk et al., 1985). The alerting effect of a nap probably influences results (Gillberg et al., 1996; Helmus et al., 1997; Tamaki et al., 2000) and can explain the improvement that has been found after the drop in performance in the afternoon. Therefore, it might be that vigilance declines continuously over the day in narcoleptic patients if naps are not permitted. In this case, the homeostatic sleep drive is likely to have increased, a view that has been suggested before (George et al., 1996). An increased influence of homeostatic processes is not a unique phenomenon, as it has also been found in healthy men who were older than 70 years (Carrier and Monk, 2000).

Most studies mentioned above measured vigilance by means of task performance or subjective scales. We explored diurnal variation in vigilance by means of EEG measures. EEG was measured both during rest and during task performance of two tasks: the SART and the memory-SART. The SART is a short vigilance task (Robertson et al., 1997) and the memory-SART is a working memory version of the SART. We included this difficult working memory task, as highly demanding tasks appear to be most sensitive in assessing vigilance (Smit et al., 2004a,b). We hypothesized that not only alpha power would be affected, but especially the lower bands in the EEG would increase. Furthermore, we predicted that power in the lower frequencies would increase throughout the day, as the sleep drive grows continuously over the waking day.

Method

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Participants

Participants were narcoleptic patients and healthy controls, who were recruited by means of advertisements. Participants in the narcolepsy group were four men and 13 women, with a mean age of 41 years (19–59), formally diagnosed with narcolepsy (see Table 1 for overview on neurological records and symptoms). Five narcoleptic patients were not on any form of medication; nine patients used ‘Modiodal’ (modafinil) and three patients used ‘Ritalin’ (methylphenidate). Patients did not take any medication or any other stimulants, such as caffeine, on the day of experimental testing. Participants in the control group were nine men and 10 women, with a comparable mean age of 40 years (23–60), who were also matched for education. All participants were (otherwise) in good health and signed an informed consent. The experiment was approved by the University Committee on Research Involving Human Subjects, CMO number: 2001/013.

Table 1.  Symptoms and HLA records of narcoleptic patients included in the study
PatientHypersomniaCataplexyHypnagogic hallucinationsSleep paralysisMSLTHLA
  1. MSLT, Multiple sleep latency test; HLA, human leukocyte antigen. +/−: atypical result; pos, positive; neg, negative.

 1++++posDR-2 pos
 2++posDR-2 pos
 3++/−++/−DR-2 pos DQB1*0602
 4+++++/−DR-2 pos
 5++negDR-2 neg/DQB1*0602 neg
 6++++posDR-2 pos
 7+++posDR-2 pos
 8++++posDQB1*0602 pos
 9+++posDQB1*0602 pos
10++++posDQB1*0602 pos
11++posDR-2 pos
12+++posDQB1*0602 pos
13++posDQB1*0602 pos
14++++posDR-2 pos
15++++posDQB1*0602 pos
16++++posDQB1*0602 pos
17+++posDR-2 pos

Design and procedure

A few days before the actual testing day, participants came to the laboratory for several hours to get acquainted with the experimental setting and to practice the tasks. This was done in order to minimize practice and familiarization effects during the testing day. On the testing day participants were not allowed to take medication. All participants came to the laboratory at 9:00 hours. Then the electrodes were placed. Directly afterwards, the first block of testing began. Testing sessions were repeated at 11:00, 13:00, 15:00, and 17:00 hours. Each testing block consisted of the following parts: EEG recording during rest (2 min with eyes closed); and EEG recording during two reaction time tasks: the SART and the memory-SART (a difficult working memory version of the SART). The order of the tests in each block was counterbalanced over participants (both in the control and the patient group). Between testing sessions, participants were allowed to go for short walks in the surroundings of the laboratory and relax in a separate room, where they could play games and read magazines. Two experimenters were constantly present in the experimenting rooms. Participants were not allowed to nap and were awoken immediately if they did tend to fall asleep.

Recording

Electrodes were placed at Fz, Cz, and Pz (Jasper, 1958). EEG was registered with Ag–Cl electrodes. The ground electrode was placed on the forehead and the left mastoid served as reference. Electro-ocular activity was recorded next to and above the right eye. Electrode impedance was <5 kΩ. Signals were band-pass filtered between 0.16 and 100 Hz and recorded digitally (512 Hz sample frequency). The EEG was checked off-line for artifacts. Epochs with eye blinks and slow eye movements were all excluded from analysis. The spectral content of the EEG was determined by fast Fourier transformations (FFTs) (frequency resolution of 0.5 Hz) with Hanning correction. For each EEG measurement, spectral content was computed for 60 epochs of 2 s. Subsequently, one grand average was made of the 60 spectral power values. We distinguished five frequency bands: delta (0.5–3.5 Hz), theta (4–7.5 Hz), alpha (8–12.5 Hz), beta1 (13–22.5 Hz), and beta2 (23–30 Hz). Power within bands was averaged.

Materials and stimulus presentation

SART

In this test, 225 single digits [25 of each of the nine digits (digits 1–9 repeated 25 times each); font: courier and sizes: 26, 28, 36 and 72] were presented visually on a computer screen (‘17 inch’: diameter of 40 cm) over a 4.3-min period. Each digit was presented for 250 ms, followed by a fixation cross that had 900-ms duration. Subjects were requested to respond with a key press (with their preferred index finger) to each digit, except in those cases the digit 3 appeared. Then they had to withhold a response. The target digit 3 and the sizes of the digits were distributed throughout the 225 trials in a prefixed quasi-random fashion, in such a way that identical digits were not clustered. The period from digit onset to digit onset was 1150 ms. Subjects were asked to give equal importance to accuracy and speed in doing the task. Both the digits and the fixation cross were presented centrally in white against a black background of the computer screen. The target digit 3 appeared 25 times in total.

Memory-SART

The same stimuli were presented as in the SART. The instruction, however, was different. Participants were asked to respond with a key press (with their preferred index finger) on each occasion a digit appeared that was the same digit as the preceding, minus 1. Thus, the digit 7 was considered to be a target if it was preceded by the digit 8. Targets appeared 26 times in total.

Data analysis

Due to many artifacts (especially eye movements), data of some participants were excluded for some testing times. If there was only one testing time missing for a particular participant, the missing value was replaced by its group mean. This was done for 15 missing values (NB: for each participant there were 15 data points for analysis: five testing times, for Fz, Cz, and Pz). Only three of the 17 participants in the narcolepsy group were unable to stay awake all the time between 15:00 and 17:00 hours, but their data were included in the analyses. These naps, with a duration of 30–40 min, occurred during the breaks. Analyses without the data of these three patients did not yield significantly different results.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

FFTs

EEG was measured during rest and during task performance (SART and memory-SART). Data were first analyzed with a 5 × 3 (time × electrode) repeated measures ANOVA, with Group as between factor. Due to interactions with electrode, separate analyses were done per electrode site. For ANOVA results, see Table 2. Spectral power figures of the FFTs are depicted in Fig. 1 for the narcolepsy group and in Fig. 2 for the control group.

Table 2.  ANOVA effects for the different frequency bands, during rest, the SART, and the memory-SART
FFTBandEffectElectrodesP-value
During restdeltaPatient > controlFz, Cz, Pz0.021 > P > 0.002
Time × groupFz, Cz, Pz0.081 > P > 0.035
theta17:00 hours > all other timesFz0.060 > P > 0.004
9:00 hours < all other timesPz0.17 > P > 0.052
Time × groupFz, PzP = 0.039; P = 0.079
alphaControl > patientFz, Cz, Pz0.027 > P > 0.008
beta113:00 hours > 9:00 hoursFzP = 0.073
During SARTtheta9:00 hours < 15:00 hours, 17:00 hoursFzP = 0.098; P = 0.022
9:00 hours < 15:00 hoursCzP = 0.092
9:00 hours < 15:00 hours, 17:00 hoursPzP = 0.009; P = 0.061
11:00 hours < 15:00 hoursPzP = 0.058
Time × groupPzP = 1.00
alpha9:00 hours < 15:00 hoursFz, Cz, Pz0.15 > P > 0.039
During memory-SARTdeltaTime × groupFzP = 0.016
theta9:00 hours < 11:00 hoursCzP = 0.013
9:00 hours < 11:00 hours, 15:00 hours, 17:00 hoursPz0.046 > P > 0.031
alphaTime × groupPzP = 0.022
beta19:00 hours < 15:00 hoursCzP = 0.048
9:00 hours < 15:00 hoursPzP = 0.077
image

Figure 1. Theta power during rest with eyes closed, in the narcolepsy and control group, at all testing times. 1 = 9:00 hours, 2 = 11:00 hours, 3 = 13:00 hours, 4 = 15:00 hours, 5 = 17:00 hours.

Download figure to PowerPoint

image

Figure 2. Alpha power during rest with eyes closed, in the narcolepsy and control group, at all testing times. 1 = 9:00 hours, 2 = 11:00 hours, 3 = 13:00 hours, 4 = 15:00 hours, 5 = 17:00 hours.

Download figure to PowerPoint

FFT during rest: delta band. Delta power was greater in the patient than in the control group at all three sites. There were time × group interactions; additional analyses per group showed that there was no Time effect at Fz in the control group, but there was a Time effect in the patient group, F(4,6) = 6.59; P = 0.022: Delta power tended to be greater at 15:00 hours than at 9:00 hours (P = 0.058). At Fz there was more delta power in the narcolepsy group than in the control group at 13:00, 15:00, and 17:00 hours (0.060 > P > 0.001); at Cz there was more delta power in the narcolepsy group at all testing times, except 9:00 hours (0.042 > P > 0.001), at Pz there was more delta power in the narcolepsy group at 13:00 hours (P = 0.011) and at 17:00 hours (P = 0.001). In sum, delta power was greater in the patient group than in the control group, from 11:00 hours onwards. There was a continuous increase in delta power in the patient group only.

FFT during rest: theta band. At Fz there was more theta power at 17:00 hours than at all other times; at Pz theta power tended to be lower at 9:00 hours than at all other times. There were time × group interactions. Additional analyses per group revealed that theta power was greater in the patient group than the control group at 13:00 and 17:00 hours at all three electrode sites (0.094 > P > 0.037). In sum, theta power increased during the day, but especially in the patient group. There was more theta power in the FFT of the narcolepsy group than in that of the control group, from 13:00 hours onwards.

FFT during rest: alpha band. There was more alpha power at all three sites in the control than in the patient group. This was the case at all testing times (0.035 > P > 0.0005).

FFT during rest: beta1 band. There were no time × group interactions, but for all three sites there appeared to be an increase in beta1 power from 15:00 to 17:00 hours in the patient group, whereas at this time there appeared to be a decrease in beta1 power in the control group (see Fig. 3).

image

Figure 3. Beta1 power during rest with eyes closed, in the narcolepsy and control group, at all testing times. 1 = 9:00 hours, 2 = 11:00 hours, 3 = 13:00 hours, 4 = 15:00 hours, 5 = 17:00 hours.

Download figure to PowerPoint

FFT during rest: beta2 band. No effects were found. However, there appeared to be the same pattern of effects as for beta1 power (see Fig. 3).

FFT during SART: delta band. No effects were found.

FFT during SART: theta band. At Fz, Cz, and Pz theta power was less at 9:00 hours than at 15:00 hours; at Pz theta power was less at 9:00 hours than 17:00 hours and there tended to be less theta power at 11:00 hours than at 15:00 hours. Furthermore, at Pz there was a trend toward a group × time interaction. Analyses per group showed that the Time effect was only significant in the patient group, F(4,4) = 6.44; P = 0.049: theta power was greater at 15:00 hours than at 9:00 hours (P = 0.050) and 11:00 hours (P = 0.042). In sum, there was a continuous increase in theta power throughout the day, which was most robust in the patient group.

FFT during SART: alpha band. Alpha power was greater at 15:00 hours than at 9:00 hours.

FFT during SART: beta1 band. No effects were present.

FFT during SART: beta2 band. There was a trend toward a time × group interaction at Fz, F(4,11) = 2.58; P = 0.096. Additional analyses revealed a nearly significant effect of Group, F(5,10) = 3.09; P = 0.061, but at no testing time did a group difference reach significance. The trend, however, was that patients showed more beta2 power than control participants, except at 17:00 hours (see Fig. 4). No significant effects were found at Cz and Pz, but the same trend was present.

image

Figure 4. Beta2 power during rest with eyes closed, in the narcolepsy and control group, at all testing times. Time 1 = 9:00 hours, Time 2 = 11:00 hours, Time 3 = 13:00 hours, Time 4 = 15:00 hours, Time 5 = 17:00 hours.

Download figure to PowerPoint

FFT during memory-SART: delta band. There was a group × time interaction at Fz, F(4,9) = 5.46; P = 0.016. Additional analyses did not reveal any significant differences between groups, due to lack of power (there were too few participants left for analysis). The trend, however, was that delta power increased continuously over the day for the patient group. The control group showed a trough in delta power at 13:00 hours. No significant effects were present at Cz and Pz, but the same trends are visible (see Fig. 5).

image

Figure 5. Delta power during rest with eyes closed, in the narcolepsy and control group, at all testing times. 1 = 9:00 hours, 2 = 11:00 hours, 3 = 13:00 hours, 4 =  15:00 hours, 5 = 17:00 hours.

Download figure to PowerPoint

FFT during memory-SART: theta band. At Cz and Pz there was less theta power at 9:00 hours than at 11:00 hours (P = 0.031); at Pz there was also less power at 15:00 and 17:00 hours.

FFT during memory-SART: alpha band. There was a group × time interaction at Pz. Additional analysis revealed a group difference, F(5,8) = 3.91; P = 0.043, but this difference did not reach significance at any particular testing time.

FFT during memory-SART: beta1 band. At Cz and Pz beta1 power was higher at 15:00 hours than at 9:00 hours.

FFT during memory-SART: beta2 band. Effects were not significant (P-values > 0.1), but there appeared to be more beta2 power in the patient group at all sites (see Fig. 5).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Alpha power during rest was reduced in the narcolepsy group. This result is in line with findings of others and has been associated with vigilance lowering (Honma et al., 2000). Lowered alpha power has been observed in conditions of very low vigilance in the presleep period (Ogilvie et al., 1991) and suggestions have been made that alpha power is linked to the input of information (Klimesch, 1999). The fact that alpha power was already lowered at 9:00 hours indicates that the spectral content is deviant at all times in narcoleptic patients and this is in line with the idea that vigilance is reduced in narcoleptic patients (George et al., 1996). Power in the alpha range was low in the narcolepsy group, but fairly stable over the day. This result indicates that alpha power is not related to the homeostatic sleep drive, which corroborates findings of others (Aeschbach et al., 1999).

Delta and theta power were enhanced, both at rest and during task performance. An increase in low frequency bands is considered to reflect low levels of vigilance (Ballard, 1996; Ogilvie et al., 1991; Paus et al., 1997). Delta power increases are usually found in sleep (Ogilvie et al., 1991), which suggests that participants may have been asleep during the 2-min recording, despite the explicit instruction to stay awake. However, after sleep onset all frequencies, including alpha and higher frequencies, are known to increase in power in healthy individuals (Ogilvie et al., 1991). Furthermore, delta power was also enhanced during task performance and patients did continue to respond, which indicates that they were not asleep. Alternatively, the EEG of narcoleptic patients might not be similar to that of healthy individuals. As a matter of fact, a slowing of EEG, combined with a decrease in alpha power, has been reported in conditions of encephalopathy (Bradshaw et al., 2001; Kamei et al., 1999). However, in these cases higher frequencies (beta) are decreased, and this was not the case in the narcolepsy group. In all, the increase in lower frequencies in the EEG of the narcolepsy group suggests a vast decline in vigilance. It is not clear whether the pattern found reflects extremely low vigilance, or a pathological brain state.

In the narcolepsy group power in delta and theta was not only enhanced, but also grew continuously throughout the day. Additionally, we found no evidence for the suggested greater postlunch dip (Pollak et al., 1992). This lack of a dip may well have been caused by the fact that patients were not allowed to nap. It thus appears that the presence of naps may indeed have confounded results in other studies (Aeschbach et al., 1999; Lafrance and Dumont, 2000; Monk et al., 1997). The present effects in delta and theta power corroborate the view that the homeostatic sleep drive is enhanced in narcolepsy (Ferri et al., 1999). In fact, theta power has been associated with the homeostatic sleep drive in healthy participants as well (Aeschbach et al., 1999). During the SART, alpha power was greatest at 15:00 hours. This result is compatible with results of Aeschbach et al. (1999), who suggested that alpha power is modulated mainly by circadian processes. The fact that this alpha effect was present in all participants again suggests that in narcolepsy circadian processes are less disturbed than homeostatic processes.

We realize that a medication effect of modafinil and methylphenidate may have influenced the results. The last dose of modafinil (elimination half-life 10–12 h) was taken 19 h before the start of the experiment. The last dose of methylphenidate (elimination half-life 2–3.5 h) was taken 13 h before the start of the experiment. However, effects of the medication on the results would have been contrary to the difference between groups that we found. Therefore, any effect must have weakened the findings in the narcolepsy group.

There was a trend toward increases in beta1 and beta2 at the end of the day. A similar combination of effects (increase in theta and beta power) has been reported in participants who were mentally fatigued and awake (Kiroy et al., 1996; Smit et al., in press). The enhancement of higher frequencies might reflect compensational processes. In a similar fashion, beta power has been linked to ‘effort to maintain wakefulness’ in participants that were sleep deprived (Corsi-Cabrera et al., 1996).

In conclusion, the spectral content of the EEG strongly suggest that narcoleptic patients suffer from a lowered level of vigilance at all testing times, both in rest and during task performance. Therefore, vigilance in medication-free narcoleptic patients does not appear to reach normal levels at any time. The steady increase in low frequency power in the EEG indicates that vigilance continuously declined further over the day in the narcolepsy group. This is in line with the idea that the homeostatic sleep drive is increased. The increase in high frequency bands, which was present during task performance at all times and during rest at the end of the day, suggests that narcoleptic patients may actively try to counteract their low level of vigilance.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We would like to thank Yvonne van de Leemput and Vincent de Groot for assistance during the experimental phase. We are grateful for the technical guidance of Jos Wittebrood and Gerard van Ooijen. We thank Hubert Voogd for constructing software regarding the reaction time tasks, and Philip van den Broek for his help with respect to EEG analyses. We thank Prof. Machiel Zwarts for his commentary concerning pathological EEG.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Achermann, P. and Borbély, A. A. Simulation of daytime vigilance by the additive interaction of a homeostatic and a circadian process. Biol. Cybernetics, 1994, 71: 115121.
  • Aeschbach, D., Matthews, J. R., Postolache, T. T., Jackson, M. A., Giesen, H. A. and Wehr, T. A. Two circadian rhythms in the human electroencephalogram during wakefulness. Am. J. Physiol., 1999, 277: R1771R1779.
  • Aguirre, M., Broughton, R. and Stuss, D. Does memory impairment exist in Narcolepsy-Cataplexy? J. Clin. Exp. Neuropsychol., 1985, 7: 1424.
  • Ballard, J. C. Computerized assessment of sustained attention: a review of factors affecting vigilance performance. J. Clin. Exp. Neuropsychol., 1996, 18: 843863.
  • Borbély, A. A. A two-process model of sleep regulation. Hum. Neurobiol., 1982, 1: 195204.
  • Bradshaw, C. B., Davis, R. L., Shrimpton, A. E., Holohan, P. D., Rea, C. B., Fieglin, D., Kent, P. and Collins, G. H. Cognitive deficits associated with a recently reported familial neurodegenerative disease: familial encephalopathy with neuroserpin inclusion bodies. Archiv. Neurol., 2001, 58: 14291434.
  • Broughton, R., Krupa, S., Boucher, B., Rivers, M. and Mullington, J. Impaired circadian waking arousal in narcolepsy-cataplexy. Sleep Res. Online, 1998, 1: 159165.
  • Carrier, J. and Monk, T. H. Circadian rhythms of performance: new trends. Chronobiol. Int., 2000, 17: 719732.
  • Corsi-Cabrera, M., Arce, C., Ramos, J., Lorenzo, I. and Guevara, M. A. Time course of reaction time and EEG while performing a vigilance task during total sleep deprivation. Sleep, 1996, 19: 563569.
  • Daan, S., Beersma, D. G. M. and Borbély, A. A. Timing of human sleep: recovery process gated by a circadian pacemaker. Am. J. Physiol., 1984, 24: R161R178.
  • Dijk, D. J., Duffy, J. F. and Czeisler, C. A. Circadian and sleep/wake dependent aspects of subjective alertness and cognitive performance. J. Sleep Res., 1992, 1: 112117.
  • Ferri, R., Pettinato, S., Nobili, L., Billiard, M. and Ferrillo, F. Correlation dimension of EEG slow-wave activity during sleep in narcoleptic patients under bed rest conditions. Int. J. Psychophysiol., 1999, 34: 3743.
  • Findley, L. J., Suratt, P. M. and Dinges, D. F. Time-on-task decrements in ‘Steer Clear’ performance of patients with sleep apnea and narcolepsy. Sleep, 1999, 22: 804809.
  • George, C. F., Boudreau, A. C. and Smiley, A. Comparison of simulated driving performance in narcolepsy and sleep apnea patients. Sleep, 1996, 19: 711717.
  • Gillberg, M., Kecklund, G., Axelsson, J. and Åkerstedt, T. The effects of a short daytime nap after restricted night sleep. Sleep, 1996, 19: 570575.
  • Helmus, T., Rosenthal, L., Bishop, C., Roehrs, T., Syron, M. L. and Roth, T. Excessive daytime sleepiness: the alerting effects of short and long naps in narcoleptic, sleep-deprived, and alert individuals. Sleep, 1997, 20: 251257.
  • Honma, H., Kohsaka, M., Fukuda, N., Kobayashi, R., Sakakibara, S. and Koyama, T. Differences in electroencephalogram power densities between genuine narcolepsy and secondary narcolepsy. Psychiatr. Clin. Neurosci., 2000, 54: 326327.
  • Hood, B. and Bruck, D. A. Comparison of sleep deprivation and narcolepsy in terms of complex cognitive performance and subjective sleepiness. Sleep Med., 2002, 3: 259266.
  • Jasper, H. H. The ten-twenty electrode system of the International Federation of Societies for Electroencephalography and Clinical Neurophysiology. Electroencephalogr. Clin. Neurophysiol., 1958, 10: 370375.
  • Johnson, M. P., Duffy, J. F., Dijk, D. J., Ronda, J. M., Dyal, C. M. and Czeisler, C. A. Short-term memory, alertness and performance: a reappraisal of their relationship to body temperature. J. Sleep Res., 1992, 1: 2429.
  • Kamei, S., Tanaka, N., Mastuura, M., Arakawa, Y., Kojima, T., Matsukawa, Y., Takasu, T. and Moriyama, M. Blinded, prospective, and serial evaluation by quantitative-EEG in interferon-alpha-treated hepatitis-C. Acta Neurol. Scand., 1999, 100: 2533.
  • Kiroy, V. N., Warsawskaya, L. V. and Voynov, V. B. EEG after prolonged mental activity. Int. J. Neurosci., 1996, 85: 3143.
  • Klimesch, W. EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Res. Rev., 1999, 29: 169195.
  • Lafrance, C. and Dumont, M. Diurnal variations in the waking EEG: comparisons with sleep latencies and subjective alertness. J. Sleep Res., 2000, 9: 243248.
  • Monk, T. H., Fookson, F. E., Kream, J., Moline, M. L., Pollak, C. P. and Weitzman, M. B. Circadian factors during sustained performance: background and methodology. Behav. Res. Meth. Instrum. Comp., 1985, 17: 1926.
  • Monk, T. H., Buysse, D. J., Reynolds, C. F., Berga, S. L., Jarrett, D. B., Begley, A. E. and Kupfer, D. J. Circadian rhythms in human performance and mood under constant conditions. J. Sleep Res., 1997, 6: 918.
  • Ogilvie, R. D., Simons, I. A., Kuderian, R. H. and MacDonald, T., Rustenburg, J. Behavioral, event-related potential, and EEG/FFT changes at sleep onset. Psychophysiology, 1991, 28: 5464.
  • Paus, T., Zatorre, R. J., Hofle, N. and Caramanos, Z. Time-related changes in neural systems underlying attention and arousal during the performance of an auditory vigilance task. J. Cogn. Neurosci., 1997, 9: 392408.
  • Pollak, C. P., Wagner, D. R., Moline, M. L. and Monk, T. H. Cognitive and motor performance of narcoleptic and normal subjects living in temporal isolation. Sleep, 1992, 15: 202211.
  • Robertson, I. H., Manly, T., Andrade, J., Baddeley, B. T. and Yiend, J. ‘Oops!’ Performance correlates of everyday attentional failures in traumatic brain injured and normal subjects. Neuropsychologia, 1997, 35: 747758.
  • Rogers, A. E. and Rosenberg, R. S. Test of memory in narcoleptics. Sleep, 1990, 13: 4252.
  • Smit, A. S., Eling, P. A. T. M. and Coenen, A. M. L. Mental effort affects vigilance enduringly: after-effects in EEG and behavior. Int. J. Psychophysiol., 2004a , 53: 239243.
  • Smit, A. S., Eling, P. A. T. M. and Coenen, A. M. L. Mental effort causes vigilance decrease due to resource depletion. Acta Psychol., 2004b , 115: 3542.
  • Tamaki, M., Shirota, A., Hayashi, M. and Hori, T. Restorative effects of a short afternoon nap (<30 min) in the elderly on subjective mood, performance and EEG activity. Sleep Res. Online, 2000, 3: 131139.