SEARCH

SEARCH BY CITATION

Keywords:

  • alertness;
  • error negativity;
  • error positivity;
  • error processing;
  • error-related negativity;
  • sleep deprivation;
  • sleepiness

Summary

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

The behavioral, cognitive, and psychophysiological effects of extended wakefulness are well known. As time awake increases, errors become more common and are often attributed to lapses in attention. Such lapses can be reflected in the error-related negativity (Ne/ERN), a negative electroencephalogram deflection occurring after errors and is thought to be related to error detection or response conflict. Following the Ne/ERN, a positive deflection (error positivity, Pe) is also observed and is thought to reflect further evaluation of the error. To elicit Ne/ERNs, the Eriksen Flanker Task was administered to 17 women (aged 19–45 years) at two levels of alertness (4 and 20 h awake). After extended wakefulness, participants reported being subjectively sleepier and performing worse, but showed no significant difference in subjective effort. Across alertness conditions, they reported a similar number of subjective errors which closely matched an objective analysis of the errors. The Ne/ERN was not significantly reduced by sleepiness in contrast to the Pe which was reduced. Behavioral slowing after errors was larger in the alert than in the sleepy condition. These results show that after 20 h of wakefulness, individuals are reacting to their errors. However, further evaluation of the error, and remediation of these errors may be impaired despite continued effort.


Introduction

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

Sleepiness influences how well an individual functions. Individuals may be able to fight extreme tiredness, but lapses of attention and errors are inevitable. The behavioral, cognitive, and psychophysiological effects of sleep restriction or extended wakefulness (>24 h) are well known (e.g. Dinges et al., 1997; Patrick and Gilbert, 1896; Van Dongen et al., 2003); however, more realistic lengths of wakefulness (<24 h) have not been as thoroughly studied. As time awake increases, errors become more common and are often attributed to lapses in attention. Several researchers (e.g. Mitchell and Williamson, 2000) have reported that the number of errors made on the job begins to rise after approximately 10 or 11 h of work, although this finding has not always been consistent (see Bendak, 2003 for a review) and much research continues to focus on reasonable hours of work and optimal scheduling. However, even once work has finished, people must continue to perform many important and often hazardous tasks such as driving a vehicle. After 20 h of wakefulness, a situation encountered frequently by many people, errors can be just as devastating as those committed by professionals such as doctors enduring longer periods of sleep deprivation. The existence of increasing likelihood of performance failure during sleepiness is well documented; however, useful identification of or counter-measures for dangerous levels of sleepiness requires a deeper understanding of the mechanisms underlying such errors and adaptiveness taken in response to performance failure.

It has typically been assumed that the increase in errors made while sleepy is primarily because of lapses in attention (Lubin, 1967) or even consciousness, i.e. micro-sleeps (Koslowsky and Babkoff, 1992), in other words, the person fails to notice the stimulus requiring a response. However, other research has shown that performance failure may occur even during objectively identifiable wakefulness as determined by polysomnography (McCarthy and Waters, 1997). At what level do these performance failures occur?

Research examining sleep onset has shown that event-related potentials (ERPs) such as the P300, indicative of stimulus processing (Polich, 1986) or categorization (Verleger et al., 2005), continue to occur even after objective responding (button press) has stopped (see Cote, 2002 for a review). Therefore, some processing of a stimulus must occur even after behavioral evidence of this processing has ceased. However, lapses, or failure to respond, can occur with increased frequency well before sleep onset. Also, in addition to these errors of omission, there are also errors of commission, where the individual responds inappropriately, often quickly without adequate consideration, analysis, or adherence to the task demands. This study addresses the electrophysiological correlates of these errors.

Error negativity/error-related negativity

Approximately 60–100 ms after an erroneous response or a response in which there is some uncertainty regarding its correctness, there is a negative electroencephalogram (EEG) deflection which has been labeled error negativity (Ne, Falkenstein et al., 1990) or the Error-Related Negativity (ERN, Gehring et al., 1993). It is most easily seen using response-locked averages from tasks that are simple but speeded so as to produce numerous errors (about 10–20%). The Ne/ERN is thought to be generated in the anterior cingulate cortex (ACC) (Stemmer et al., 2003; Van Veen and Carter, 2002). The Ne/ERN has been hypothesized to reflect either error detection (Holroyd et al., 2002) or response conflict (Yeung et al., 2004).

Regardless of which account is most accurate, the Ne/ERN appears to play some role in behavioral modification. It has been shown that the amplitude of the Ne/ERN correlates with the subjective confidence of the response; the Ne/ERN is larger after trials where the participant is most certain an error had been committed (Scheffers and Coles, 2000). Gehring et al. (1993) also found that the amplitude of the Ne/ERN correlates positively with the increase in response time on the following trial. In other words, the larger the Ne/ERN, the greater the behavioral compensation after errors.

The Ne/ERN has also been shown to be sensitive to several other manipulations and individual differences. For example, for some individuals, the Ne/ERN varies with the importance of the error (Pailing and Segalowitz, 2004a). Individuals who produce smaller Ne/ERNs tend to make impulsive errors (Pailing et al., 2002) and when given instructions to focus on speed over accuracy, people in general produce a smaller Ne/ERN to an incorrect response (Gehring et al., 1993; Ullsperger and Szymanowski, 2004)). It has been reported that conscious recognition of the error may not be required for the Ne/ERN to be observed (Nieuwenhuis et al., 2001); however, the research that links Ne/ERN amplitude with conscious evaluation of error probability (Scheffers and Coles, 2000) or degree of behavioral change (Gehring et al., 1993) suggests that Ne/ERN is associated with conscious error recognition.

Ne/ERN and lowered arousal because of sleepiness or alcohol consumption

Scheffers et al. (1999) reported a reduction in the Ne/ERN after 24 h of wakefulness and this effect was attributed to a failure to detect errors. However, the participants were tested a total of nine times, in addition to practice sessions, so there may have been an effect because of the multiple testing sessions. Perhaps familiarity with the task may have interacted with sleepiness such that what is being reflected is an increased boredom or habituation factor rather than a drop in error detection. Alcohol intake can also affect the Ne/ERN. Ridderinkhof et al. (2002) found that even a moderate dose of alcohol (Blood Alcohol Concentration (BAC) ∼0.04%) reduced the amplitude of the Ne/ERN by approximately one-third, but no further reductions were seen at higher doses (BAC ∼0.10%). Posterror slowing, a classic behavioral outcome associated with compensatory actions did not occur at either level of intoxication.

Ne/ERN and motivation

The Ne/ERN and Pe have been shown to be sensitive to both internal (e.g. personality) and external (e.g. reward or instruction) variables. The Ne/ERN is larger in individuals with obsessive–compulsive disorder (Gehring et al., 2000) and smaller in some contexts in individuals with a low level of socialization (Dikman and Allen, 2000; Santesso et al., 2005). Hence, those people who are more concerned with performance or with what others observe about them appear to produce larger Ne/ERNs. The Ne/ERN has also been shown to be sensitive to instruction. In a task involving two stimulus factors, the size of the Ne/ERN was influenced by the amount of reward associated with the stimulus factor incorrectly responded to, although this varied with personality factors related to conscientiousness (Pailing and Segalowitz, 2004b). Therefore, the Ne/ERN appears to depend on the motivational state.

Error positivity

Following the Ne/ERN, a positive deflection (error positivity, Pe, Falkenstein et al., 1990) has been observed, but is less studied. The Pe is typically maximal at parietal sites, and reaches maximum amplitude between 200 and 400 ms after an erroneous response. Because of similar polarity, topography and latency, the Pe may be a P3b related to the error (as opposed to the stimulus) and research has shown that it is likely related to error recognition, response strategy or subjective error processing (see Falkenstein, 2004, for review). Scheffers et al. (1999) did not examine the Pe so the effect of sleepiness on the Pe remains unknown.

The current study examines the effect of sleepiness on the Ne/ERN and the Pe. We employed a task known to produce the Ne/ERN and Pe: the Eriksen Flanker task (Eriksen and Eriksen, 1974). In addition we also asked participants to evaluate their subjective sleepiness, effort, and performance immediately after the task was completed.

Method

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

Subjects

Seventeen females (aged 19–45 years) participated. They were free of previous head injury, serious sleep disorders, excessive daytime sleepiness, neurological problems, and medications which may affect alertness (all measured by questionnaire and interview). They scored in the mid-range on the morningness/eveningness scale (Horne and Ostberg, 1976) as well as the Epworth Sleepiness Scale (ESS) (John, 1991) and had typical sleep patterns, defined as a typical bedtime of 22:00–0:00 hours and a rising time of 6:00–8:00 hours. Participants completing the entire protocol were paid an honorarium for their participation.

Procedure

Participants were given a sleep log to complete for 2 weeks to assess their typical sleep and activity patterns, as well as food, caffeine and alcohol intake. Participants were instructed not to use alcohol or caffeine the night before or day of testing, to sleep 7–8 h the night before testing, awaken at their normal time and not nap the day of testing.

Participants were tested three times. During the first session, the participant was introduced to the lab and all procedures were explained and an abbreviated version of each test was administered. This session was meant to familiarize participants with the tests and reduce any novelty or arousing effect of the situation and/or tests being administered.

The two experimental sessions used identical procedures and were counterbalanced across participants. The alert session began approximately 2 h after the participant awakened (i.e. between 09:00–10:00 hours) and the sleepy session began approximately 3 h after the participants normal bedtime (approximately 02:00–03:00 hours).

Participants arrived at the lab 1 h before each testing session for electrode hookup. An electrode cap (Electro-Cap International, Inc., Eaton, OH, USA) containing 44 tin electrodes embedded within the material of the cap was applied. Electrodes were also placed on each earlobe, the outer canthus of both eyes, and supraorbital ridge of the right eye. All EEG and electrooculogram (EOG) electrodes as well as the left ear were referenced to the right ear and a mastoid electrode served as ground. EOG for this task was recorded using a bi-polar reference (outer canthus and supraorbital ridge of the right eye). EEG electrodes were re-referenced offline to an average of the two ears. All electrode impedances were maintained below 5 kΩ. EEG and EOG data were converted from analog to digital signals with a window of ±250 μV (12-bit resolution) using a sampling frequency of 512 Hz and band pass frequency of 0.16–100 Hz. The EEG signal was amplified at a gain of 10 000 using a Sensorium Inc. amplifier system (Sensorium Inc., RR#3, Box 3275, Charlotte, VT, USA). instep (a commercial data acquisition program) was used to present stimuli and acquire data. To eliminate electrical interference, EEG was further filtered offline using a 60 Hz notch filter.

Flanker task

The present paper presents results from the Flanker task, part of a larger battery administered in this study. The Flanker task took place approximately 1 h into each 2-h testing session. Actual mean time awake before testing was 4.0 h (SD = 0.8) in the alert session and 19.9 h (SD = 1.2) in the sleepy session. The Flanker task requires participants to respond quickly (Inter-trial interval (ITI) = 1250 ms) via a forced choice key press (counterbalanced across participants and conditions) to central target letters (H or S) from both congruent (HHHHH, SSSSS, n = 160) and incongruent (HHSHH, SSHSS, n = 320) arrays. They were instructed that the letters would appear rapidly so they must react quickly to perform the task; however, it was emphasized that speed and accuracy were equally important.

Subjective measures

Visual Analog Scales (VAS) were used to measure subjective estimates of sleepiness, effort and performance. Each VAS was a 10 cm line with two anchor points indicating polar opposites on each dimension. The anchors used for each scale were as follows: sleepiness (very sleepy–very alert), effort (no effort–maximum effort), and performance (very poorly–very well). These measurements were taken immediately after each task was completed. In addition each participant was asked to estimate the number of errors made on the Flanker task.

Ne/ERN and Pe measurement

The effects of eye movements on scalp EEG were partialed out using a regression procedure that displays the residual scalp ERP with the bipolar vertical eye channel signal removed on a trial-to-trial basis, permitting manual rejection of the trial in the rare case when there is overcorrection. This method thus reduces noise introduced by overcorrection that is occasionally found in automated eye correction procedures. After this correction, correct and error ERPs were created based on response-locked averages and smoothed with a five-point moving window average. Response-locked averages still preserve the P300 to the stimulus (Verleger et al., 2005), normally within the 200 ms preceding the response, as can be seen in Fig. 1. The amplitude of the P300 is typically reduced by sleepiness (Lee et al., 2004), and therefore, to use the pre-response period as a baseline would confound the P300 amplitude effects with the Ne/ERN effects. Thus, to minimize any effect that the stimulus P300 had on these averages, a baseline of −600 to −400 ms relative to response was used. The amplitude of the Ne/ERN was measured as the maximum deflection between 50 and 120 ms postresponse. As the Pe often had slow rising waveforms or poorly defined peaks, Pe amplitude was measured as the mean amplitude between 200–400 ms (Nieuwenhuis et al., 2001). The sites scored for analyses were FCz for the Ne/ERN, and Pz for the Pe. These are the sites of maximal deflection for these two components.

image

Figure 1. Response-locked electroencephalogram averages (Ne/ERN and Pe marked) to correct responses and errors in the Flanker task in both the alert and sleepy conditions. The similar sized Ne/ERN (P = 0.50) but reduced Pe can be seen (P < 0.001). (Ne, error negativity; ERN, error-related negativity; Pe, error positivity).

Download figure to PowerPoint

Results

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

Two participants were excluded from all EEG analyses because of technical problems. The recording parameters were set incorrectly for one, and for the other, the original data file was corrupted. Only the 12 participants who showed a clearly formed negative deflection in the appropriate time frame were included in analysis of the Ne/ERN. In all other analyses the maximum number of available data points was used.

Subjective and objective behavioral measures

Paired t-tests were conducted to assess the subjective and objective measures across conditions. During the sleepy condition, participants reported being more sleepy (M = 9.07, SD = 8.00 versus M = 49.00, SD = 25.99) [t(13) = 6.07, P < 0.001], and performing worse (M = 26.63, SD = 15.70 versus M = 44.50, SD = 19.62) [t(13) = 3.04, P < 0.01], even though neither the subjective estimate of the number of errors (M = 24.00, SD = 11.98 versus M = 24.08, SD = 10.80) [t(12) = 0.02, P = 0.99] nor the objective number of errors made (M = 28, SD = 17.62 versus M = 26.12 SD = 12.04) [t(16) = 0.47, P = 0.64] was significantly different across conditions. There was no significant difference in subjective effort across conditions (M = 73.71, SD = 18.74 versus M = 71.29, SD = 14.53) [t(13) = 0.54, P = 0.60].

Error-related negativity

Paired t-tests showed no significant difference in Ne/ERN amplitude measured at FCz across conditions (M = −8.22, SD = 4.68 versus M = −7.04, SD = 5.10) [t(11) = 0.69, P = 0.50] (see Fig. 1). There appears to be a peak-to-peak (P300-Ne/ERN) difference in the amplitudes; however, this difference is only a trend (M = 11.08, SD = 4.69; M = 13.98, SD = 4.47), [t(11) = 1.94, P = 0.08] and is because of a significant difference in the positive peak (P300) just prior to the Ne/ERN [t(11) = 2.65, P < 0.05]. There were no significant correlations between the amplitude of the Ne/ERN and subjective number of errors, objective number of errors, subjective effort, or performance in either condition.

Error positivity

The Pe was significantly reduced in the sleepy condition (M = 4.24, SD = 3.15; M = 7.50, SD = 5.19), [t(14) = 3.44, P = 0.004]. It has been reported that the number of errors may influence the amplitude of the Pe (Dywan et al., 2004); however, the Pe did not correlate with the subjective number of errors, objective number of errors, subjective effort, or performance in either condition. It is, therefore, unlikely the number of errors (real or perceived) influenced the amplitude of the Pe.

Behavioral corrections after errors

After erroneous responses, participants typically slow down their response to the following trial to reduce the probability of making another error (Gehring et al., 1993; Hajcak et al., 2003). To assess the behavioral effects of errors in this experiment, a two (alert/sleepy) by two (correct/error) within-subject anova was conducted using response time on correct trials following errors and following correct responses as the dependent variable. Behavioral slowing after errors was more pronounced in the alert condition (see Fig. 2). Response times on correct trials following errors slowed by 28.5 ms when alert, but only by 9.1 ms when sleepy [F(1,14) = 7.71, P = 0.015].

image

Figure 2. Mean response time for correct trials following correct (light bars) and correct trials following error trials (dark bars) in both alert and sleepy conditions. The interaction can been seen showing that there is relatively less compensatory slowing after errors when one is sleepy (P < 0.05).

Download figure to PowerPoint

Discussion

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

This experiment was designed to assess the effect of moderate sleep deprivation on error processing. After 20 h of wakefulness, individuals continue to react electrophysiologically to errors. This can be seen in the similarity in the Ne/ERN across both levels of alertness. However, posterror evaluation (reduced Pe), and remediation of these errors (behavioral data) were impaired despite participants reporting sustained effort.

Scheffers et al. (1999) have reported a reduction in the Ne/ERN after 24 h of wakefulness. However, they used memory and visual search tasks with multiple levels (three and six items) which are not typical Ne/ERN paradigms. Although they argued that the reductions in Ne/ERN amplitude (approx. 3 μV) are primarily because of time awake, response times and error rates varied based on type of task, task load, time on task, as well as time awake. For example, the amplitude of the Ne/ERN was significantly larger in the memory search and significantly smaller with a larger memory load (six items). Also, the response times and error rates were significantly greater in the visual search task. Participants in the experiment conducted by Scheffers et al. (1999) were also tested multiple times. Therefore, the conclusion that reduction in the Ne/ERN is because of time awake and more specifically a ‘decrease in the quality of perceptual processing’ may not be warranted.

Similar to our findings, Scheffers et al. (1999) did not find evidence of behavioral slowing after errors in either task (P = 0.23 and 0.28) although this is likely because of different types of task employed or may have been an issue of power with only eight participants. Thus, other factors may have attenuated the Ne/ERN in the participants when sleepy, such as being less certain of their performance (Coles et al., 2001; Pailing and Segalowitz, 2004a) or being less motivated to perform well (Pailing and Segalowitz, 2004b). Scheffers et al. (1999) also had only male participants and the present study used only females, so gender differences also cannot be ruled out.

Ridderinkhof et al. (2002) found that after alcohol consumption, the Ne/ERN amplitude was significantly reduced by approximately 2.1 μV. This reduction in Ne/ERN amplitude after alcohol ingestion was also related to a failure to adjust behavioral responses after errors, even after the error rate was controlled for, and these results were interpreted as a breakdown in the recognition of errors. In contrast, the present experiment found a non-significant reduction in the Ne/ERN of approximately 1.2 μV, with a significant reduction in the Pe and a reduction in posterror slowing.

Ridderinkhof et al. (2002) interpreted their findings as indicating that alcohol impairs performance monitoring, both at an electrophysiological level of activity in the ACC and behaviorally with respect to posterror slowing. Our results suggest a somewhat different pattern: participants when sleepy can monitor errors at the level of the ACC but they do not adjust behavior to compensate for these errors, nor do they respond as much with the attentional–emotional response reflected in the Pe. Thus, while alcohol deteriorates performance monitoring, sleepiness reduces the motivation to adapt behavior even when errors are detected. Further investigation into the combined effects of alcohol and sleepiness appears warranted. If alcohol reduces awareness of errors and sleepiness reduces evaluation, then a combination may prove more dangerous than simply the additive effects of both conditions.

Previous research has shown that reductions in the Pe may be related to the number of errors made (Dywan et al., 2004; Falkenstein, 2004); however, we found that the correlation between number of errors and size of the Ne/ERN or Pe was not significant nor in fact did performance deteriorate with sleepiness. Therefore, the reduction in the Pe cannot be attributed to an increase in errors. However, despite no significant reduction in the subjective estimate of the number of errors or the objective assessment of number of errors, participants did report a subjective decline in performance. This supports Falkenstein's (2004) hypothesis that the Pe may reflect further subjective/emotional assessment of errors.

We found diminished behavioral reactivity to errors. This indicated that these participants were failing to alter their behavior (increase response time after error sufficiently) in order to compensate for their impairment because of sleepiness and improve their performance. If they perceived their performance as worse and were trying to compensate, the increase in response time following errors when sleepy should have been even greater than the increase when alert, but this was not the case.

In light of the fact that we failed to find any significant difference in effort or Ne/ERN amplitude across conditions, we conclude that although awareness (they notice) and motivation (they care) remain relatively intact, error evaluation (Pe) is impaired after extended wakefulness. This has very serious implications for anyone who find themselves in the situation of having to perform a dangerous task while sleepy. We chose the time frame of approximately 20 h awake because this represents an amount of wakefulness (or sleep deprivation) commonly encountered by a large proportion of the population at some point in their lives. Whether it is a doctor on 24-h call, a member of the armed forces on extended maneuvers, a long distance professional driver, a shift worker on a double shift, or simply someone driving home after a social evening, the failure to adequately compensate for diminished alertness can have dire outcomes.

Further research is needed to assess how much wakefulness is required to produce deficits, what counter measures (if any) may reduce this impairment and what effects other situational variables (secondary or distracting tasks, alcohol consumption, time on task) and individual variables (gender, age, personality) may have on error processing while sleepy. The common countermeasure of consuming caffeine may be entirely counterproductive if it does not improve motivation and ability to compensate behavior appropriately.

Acknowledgements

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

This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada (No. 122222–03) to SJS. We would like to thank James Desjardins for his help with the data handling. We also thank our anonymous reviewers for their helpful comments and suggestions.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Bendak, S. 12-h workdays: current knowledge and future directions. Work Stress, 2003, 17: 321336.
  • Coles, M. G. H., Scheffers, M. K. and Holroyd, C. B. Why is there an ERN/Ne on correct trials? Response representations, stimulus-related components, and the theory of error-processing. Biol. Psychol., 2001, 56: 173189.
  • Cote, K. A. Probing awareness during sleep with the auditory oddball paradigm. Int. J. Psychophysiol., 2002, 46: 227241.
  • Dikman, Z. V. and Allen, J. J. B., Error monitoring during reward and avoidance learning in high- and low-socialized individuals. Psychophysiology, 2000, 37: 4354.
  • Dinges, D. F., Pack, F., Williams, K., Gillen, K. A., Powell, J. W., Ott, G. E., Aptowicz, C. and Pack, A. I. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4–5 h per night. Sleep, 1997, 20: 267.
  • Dywan, J., Mathewson, K. J. and Segalowitz, S. J. Error related ERP components and source monitoring in older and younger adults. In: M.Ullsperger and M.Falkenstein (Eds) Errors, Conflicts, and the Brain: Current Opinions on Performance Monitoring. Max Planck Institute for Cognition and Neuroscience, Leipzig, 2004: 184191.
  • Eriksen, B. A. and Eriksen, C. W. Effects of noise letters upon the identification of a target letter in a nonsearch task. Percept. Psychophys., 1974, 16: 143149.
  • Falkenstein, M. ERP correlates of erroneous performance. In: M.Ullsperger and M.Falkenstein (Eds) Errors, Conflicts, and the Brain: Current Opinions on Performance Monitoring. Max Planck Institute for Cognition and Neuroscience, Leipzig, 2004: 514.
  • Falkenstein, M., Hohnsbein, J., Hoormann, J. and Blanke, L. Effects of errors in choice reaction tasks on the ERP under focused and divided attention. In: C. M. H.Brunia, A. W. K.Gaillard and A.Kok (Eds) Psychophysiological Brain Research. Tilburg University Press, Tilburg, 1990: 192195.
  • Gehring, W. J., Goss, B., Coles, M. G. and Meyer, D. E. A neural system for error detection and compensation. Psychol. Sci., 1993, 4: 385390.
    Direct Link:
  • Gehring, W. J., Himle, J. and Nisenson, L. G. Action-monitoring dysfunction on obsessive-compulsive disorder. Psychol. Sci., 2000, 11: 16.
    Direct Link:
  • Hajcak, G., McDonald, N. and Simons, R. F. To err is autonomic: error-related brain potentials, ANS activity, and post-error compensatory behavior. Psychophysiology, 2003, 40: 895903.
  • Holroyd, C. B., Coles, M. G. H. and Nieuwenhuis, S. Medial prefrontal cortex and error potentials. Science, 2002, 296: 16101611.
  • Horne, J. A. and Ostberg, O. A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int. J. Chronobiol., 1976, 4: 97100.
  • John, M. W. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep, 1991, 14: 540545.
  • Koslowsky, M. and Babkoff, H. Meta-analysis of the relationship between total sleep deprivation and performance. Chronobiol. Int., 1992, 9: 132136.
  • Lee, H.-J., Kim, L., Kim, Y.-K., Suh, K.-Y., Han, J., Park, M.-K., Park, K.-W. and Lee, D.-H. Auditory event-related potentials and psychological changes during sleep deprivation. Neuropsychobiology, 2004, 50: 15.
  • Lubin, A. Performance under sleep loss and fatigue. In: S. S.Kety, E. V.Evarts and H. L.Willimas (Eds) Sleep and Altered States of Consciousness. Williams and Wilkins, Baltimore, 1967: 506513.
  • McCarthy, M. E. and Waters, W. F. Decreased attentional responsitivity during sleep deprivation: orienting response latency, amplitude and habituation. Sleep, 1997, 20: 114123.
  • Mitchell, R. J. and Williamson, A. M. Evaluation of an 8 h versus a 12 h shift roster on employees at a power station. Appl. Ergon., 2000, 31: 8393.
  • Nieuwenhuis, S., Ridderinkhof, K. R., Blom, J., Band, G. P. H. and Kok, A. Error-related brain potentials are differentially related to awareness of response errors: evidence from an anti-saccade task. Psychophysiology, 2001, 38: 752760.
  • Pailing, P. E. and Segalowitz, S. J. The effects of uncertainty in error monitoring on associated ERPs. Brain Cogn, 2004a, 56: 215233.
  • Pailing, P. E. and Segalowitz, S. J. The error-related negativity as a state and trait measure: motivation, personality, and ERPs in response to errors. Psychophysiology, 2004b, 41: 8495.
  • Pailing, P. E., Segalowitz, S. J., Dywan, J. and Davies, P. L. Error negativity and response control. Psychophysiology, 2002, 39: 198206.
  • Patrick, G. T. and Gilbert, J. A. On the effects of loss of sleep. Psychol. Rev., 1896, 3: 469483.
  • Polich, J. P300 development from auditory stimuli. Psychophysiology, 1986, 23: 590597.
  • Ridderinkhof, R., De Vlugt, Y., Bramlage, A., Spaan, M., Elton, M., Snel, J. and Band, G. P. H. Alcohol consumption impairs detection of performance errors in mediofrontal cortex. Science, 2002, 298: 22092211.
  • Santesso, D., Segalowitz, S. J. and Schmidt, L. A. ERP correlates of error monitoring in 10 year olds are related to socialization. Biol. Psychol., 2005, 70: 7987.
  • Scheffers, M. K. and Coles, M. G. H. Performance monitoring in a confusing world: error-related brain activity, judgments of response accuracy, and types of errors. J. Exp. Psychol. Hum. Percept. Perform., 2000, 26: 141151.
  • Scheffers, M. K., Humphrey, D. G., Stanny, R. R., Kramer, A. F. and Coles, M. G. H. Error-related processing during a period of extended wakefulness. Psychophysiology, 1999, 36: 149157.
  • Stemmer, B., Segalowitz, S. J., Witzke, W. and Schönle, P. W. Error detection in patients with lesions to the medial prefrontal cortex: an ERP study. Neuropsychologia, 2003, 42: 118130.
  • Ullsperger, M. and Szymanowski F. ERP correlates or error relevance. In: M.Ullsperger and M.Falkenstein (Eds) Errors, Conflicts, and the Brain: Current Opinions on Performance Monitoring. Max Planck Institute for Cognition and Neuroscience, Leipzig, 2004: 171177.
  • Van Dongen, H. P., Maislin, G., Mullington, J. M. and Dinges, D. F. The cumulative cost of additional wakefulness: dose–response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep, 2003, 26: 117126.
  • Van Veen, V. and Carter, C. S. The anterior cingulate as a conflict monitor: fMRI and ERP studies. Physiol. Behav., 2002, 77: 477482. (The Pittsburgh special issue).
  • Verleger, R., Jaskowski, P. and Wascher, E. Evidence for an integrative role of P3b in linking reaction to perception. J. Psychophysiol., 2005, 19: 165181.
  • Yeung, N., Botvinick, M. M. and Cohen, J. D. The neural basis of error detection: conflict monitoring and the error-related negativity. Psychol. Rev., 2004, 111: 931959.