Sleep and Executive Performance: The Hidden Impairment

Van Dongen, Maislin, Mullington and Dinges (2003, SLEEP) found that 14 days of 6 hours per night produced cognitive deficits equivalent to one full night of total sleep deprivation, while participants reported only modest subjective sleepiness. They were severely impaired and didn't know it.

2026-02-19 6 min read

The Van Dongen Study

Hans Van Dongen, Greg Maislin, Janet Mullington, and David Dinges published “The Cumulative Cost of Additional Wakefulness: Dose-Response Effects on Neurobehavioral Functions and Sleep Physiology from Chronic Sleep Restriction and Total Sleep Deprivation” in SLEEP in 2003 (Vol. 26, No. 2, pp. 117–126).

The study randomized participants to spend either 4 hours, 6 hours, or 8 hours in bed per night for 14 consecutive days. A separate group underwent total sleep deprivation. Cognitive performance was measured multiple times daily using the psychomotor vigilance task (reaction time) and digit symbol substitution (a measure of processing speed and attention).

6 hrs = 1 sleepless night

cognitive performance after 14 days of 6 hours per night was equivalent to approximately one full night of total sleep deprivation, measured by psychomotor vigilance task performance. Four hours per night for 14 days approached the performance level of two consecutive sleepless nights.

Van Dongen, Maislin, Mullington & Dinges (2003), SLEEP, 26(2)

Performance in the 4-hour and 6-hour groups declined progressively across the 14 days with no stabilization. The impairment continued to accumulate across the entire study period. There was no adaptation or tolerance to the sleep restriction. The 8-hour group showed no significant performance decline across the same period.

The Underestimation Problem

The most consequential finding in the Van Dongen study is not the performance impairment itself. It is the divergence between objective impairment and subjective awareness of it.

Participants rated their sleepiness daily using the Karolinska Sleepiness Scale. The 4-hour and 6-hour groups showed only modest increases in subjective sleepiness ratings across the 14 days. Critically, participants in these groups reported similar levels of sleepiness to each other. They did not subjectively differentiate their conditions despite the 4-hour group showing substantially worse objective performance. And neither restriction group reported the level of subjective sleepiness that matched their objective cognitive impairment.

The practical interpretation: people who chronically sleep 6 hours per night feel somewhat tired but functional. They are not experiencing the dramatic subjective impairment that would signal a problem to them. Their felt competence is relatively intact; their actual performance on attention-demanding, time-sensitive tasks is severely compromised. The person who is most impaired by their sleep pattern is the least positioned to self-diagnose it.

Executive Function and High-Stakes Decisions

The cognitive functions most impaired by sleep restriction are precisely those most required for senior leadership: sustained attention, working memory, emotional regulation, and judgment under uncertainty. The psychomotor vigilance task captures sustained attention, which is the ability to maintain performance on a demanding task over time. Sleep-restricted participants show increasing lapses (failures to respond within the allowed window) that accumulate across the study period.

Risk assessment and negotiation. Sleep-restricted individuals show reduced sensitivity to social cues, increased risk tolerance, and less accurate probabilistic reasoning. These are the functions most needed in high-stakes negotiations, investment decisions, and crisis responses, exactly the contexts in which senior leaders who sleep less tend to be most active.
Emotional regulation. Amygdala reactivity to threatening stimuli increases with sleep restriction, while prefrontal regulatory capacity decreases. The result: sleep-restricted executives tend to over-react to negative information and under-regulate conflict responses, a pattern that produces worse decisions precisely in the emotionally loaded situations that most require good judgment.
Meetings and scheduling. The timing of high-stakes meetings matters because cognitive performance tracks circadian rhythms and sleep pressure. Scheduling the most demanding cognitive work (complex decisions, adversarial negotiations, sensitive personnel matters) at times when sleep-restricted cognitive performance is at its nadir compounds the impairment. The executive who sleeps 6 hours and schedules their hardest work at the end of the day is stacking two performance penalties.

Frequently Asked Questions

Can you catch up on sleep debt over the weekend?

The research suggests partial recovery at best. Van Dongen and colleagues conducted follow-up work showing that a recovery period of several days' full sleep does restore some performance, but does not fully return to pre-restriction baseline after extended restriction. The cognitive costs of chronic partial sleep deprivation are not fully repaid by a weekend of recovery sleep. Other research has shown that self-assessed recovery (how recovered you feel) significantly exceeds objective performance recovery after catch-up sleep. The subjective sense of being restored is faster than the actual restoration of performance. The practical implication: weekend recovery sleep is better than no recovery, but does not fully erase a week of 6-hour nights.

Does sleep restriction affect some types of cognitive performance more than others?

Yes. The pattern of impairment is not uniform across cognitive functions. Sustained attention and processing speed show the largest and most consistent effects of sleep restriction. Working memory is impaired but somewhat less robustly. Higher-order executive functions (planning, reasoning, creativity) are affected but more variably, and participants' subjective assessment of these higher functions is particularly poorly calibrated. People tend to feel that their judgment and reasoning are intact while their sustained attention is measurably degraded. This is the dangerous part: the functions that feel intact under sleep restriction are the ones most relevant to senior leadership, and subjective confidence in those functions does not reliably track their actual state.

At what point does sleep deprivation become a material performance liability for executives?

The Van Dongen study suggests the performance liability begins with restriction below 8 hours, though the effect is dose-dependent: 7 hours shows smaller effects than 6, which shows smaller effects than 4. The threshold where impairment becomes consequential depends on the demands of the task: for tasks requiring sustained attention and fast accurate responding (trading, real-time negotiation, crisis management), the impairment from 6-hour restriction is measurable and significant within days. For less time-pressured cognitive work, the threshold may be somewhat lower. The more important consideration for executives is the underestimation problem: the confidence that you are performing well is not calibrated to your actual performance state, so self-monitoring is not a reliable safety mechanism.