Changes in performance and bio-mathematical model performance predictions during 45 days of sleep restriction in a simulated space mission

Abstract

Lunar habitation and exploration of space beyond low-Earth orbit will require small crews to live in isolation and confinement while maintaining a high level of performance with limited support from mission control. Astronauts only achieve approximately 6 h of sleep per night, but few studies have linked sleep deficiency in space to performance impairment. We studied crewmembers over 45 days during a simulated space mission that included 5 h of sleep opportunity on weekdays and 8 h of sleep on weekends to characterize changes in performance on the psychomotor vigilance task (PVT) and subjective fatigue ratings. We further evaluated how well bio-mathematical models designed to predict performance changes due to sleep loss compared to objective performance. We studied 20 individuals during five missions and found that objective performance, but not subjective fatigue, declined from the beginning to the end of the mission. We found that bio-mathematical models were able to predict average changes across the mission but were less sensitive at predicting individual-level performance. Our findings suggest that sleep should be prioritized in lunar crews to minimize the potential for performance errors. Bio-mathematical models may be useful for aiding crews in schedule design but not for individual-level fitness-for-duty decisions.

Figure 1

HERA habitat (https://www.nasa.gov/analogs/hera).

Figure 2

Study schematic for the 45-day protocol. Gray bars indicate schedule sleep. Triangles indicate the days and timing of psychomotor vigilance tests PVT. Hatched region indicates the duration of time when crewmembers were allowed to consume caffeine. The vertical dashed line indicates midnight.

Figure 3

Average (black and red circles) psychomotor vigilance task (PVT) performance and individual daily mean values (light gray circles) by mission day for mean reaction time (A), lapses > 500 ms (B), fastest 10% reaction time (C), response speed (D), slowest 10% reaction time (E), and Samn Perelli ratings (F) by day of mission. Note differences in y-axis scale for mean, fastest and slowest reaction times. Black circles indicate days following 5 h of sleep, red circles indicate days following 8 h of sleep. RT reaction time, ms milliseconds, error bars reflect the standard error of the mean.

Figure 4

Average psychomotor vigilance task (PVT) performance for mean reaction time (A), lapses (B), fastest 10% reaction time (C), response speed (D), slowest 10% reaction time (E), and Samn Perelli ratings (F) by prior night’s sleep duration. Note differences in y-axis scale for mean, fastest and slowest reaction times. RT reaction time, ms milliseconds. ***p < 0.01.

Figure 5

Average (black circles) psychomotor vigilance task (PVT) performance and individual (light gray circles) mean performance by session for mean reaction time (A), lapses (B), fastest 10% reaction time (C), response speed (D), and slowest 10% reaction time (E) by session of the day. Note differences in y-axis scale for mean, fastest and slowest reaction times. RT reaction time, ms milliseconds, error bars reflect the standard error of the mean.

Figure 6

Relationship between model predictions (A) and scaled model predictions (B) and associated confidence intervals with actual lapses (± standard error) by day of the mission. Actual performance measures are shown as filled circles, model predictions are shown as open symbols as follows: triangles = adenosine-circadian model, squares = unified model, diamonds = state-space model, stars = SAFTE model. Black circles indicate days following 5 h of sleep, red circles indicate days following 8 h of sleep. Note, SAFTE model outcome “cognitive effectiveness” ranges from 0–100 and is plotted in the inverse on panel (A).

Figure 7

Relationship between model predictions (A) and scaled model predictions (B) and associated confidence intervals with actual lapses (± standard error) by session of the day for all days combined. Actual performance measures are shown as filled circles, model predictions are shown as open symbols as follows: triangles = adenosine-circadian model, squares = unified model, diamonds = state-space model, stars = SAFTE model. Note, SAFTE model outcome “cognitive effectiveness” ranges from 0–100 and is plotted in the inverse on panel (A).

Figure 8

Relationship between scaled model predictions and actual lapses (± standard error) by session of the day following 8 h sleep (filled symbols) and 5 h of sleep (open circles). Actual performance measures are shown as circles (filled for 8-h, open for 5-h), model predictions are shown as follows: triangles = adenosine-circadian model (A), squares = unified model (B), diamonds = state-space model (C), stars = SAFTE model (D). Note, confidence intervals are narrow for the model predictions and are not visualized on the plot.