Attentional failures after sleep deprivation represent moments of cerebrospinal fluid flow

Abstract

Sleep deprivation rapidly disrupts cognitive function, and in the long term contributes to neurological disease. Why sleep deprivation has such profound effects on cognition is not well understood. Here, we use simultaneous fast fMRI-EEG to test how sleep deprivation modulates cognitive, neural, and fluid dynamics in the human brain. We demonstrate that after sleep deprivation, sleep-like pulsatile cerebrospinal fluid (CSF) flow events intrude into the awake state. CSF flow is coupled to attentional function, with high flow during attentional impairment. Furthermore, CSF flow is tightly orchestrated in a series of brain-body changes including broadband neuronal shifts, pupil constriction, and altered systemic physiology, pointing to a coupled system of fluid dynamics and neuromodulatory state. The timing of these dynamics is consistent with a vascular mechanism regulated by neuromodulatory state, in which CSF begins to flow outward when attention fails, and flow reverses when attention recovers. The attentional costs of sleep deprivation may thus reflect an irrepressible need for neuronal rest periods and widespread pulsatile fluid flow.

Introduction:

Sleep plays a fundamental role in maintaining brain health and preserving cognitive performance. Despite the intense biological need for sleep, sleep deprivation is a major facet of modern life. A single night of lost sleep can cause noticeable cognitive impairment, including attentional failures, in which individuals fail to properly react to an easily detected external stimulus (1–5). The behavioral deficits caused by sleep deprivation carry a major cost for organismal survival: for example, a momentary lapse in attention while driving a car can have life-threatening outcomes (6). Sleep deprivation reliably induces attentional failures despite this cost, suggesting that these deficits reflect an unavoidable need of the brain for sleep. However, the neural basis of sleep-deprivation-induced attentional failures is not yet well understood.

Acute sleep deprivation has widespread effects on both local and global aspects of neurophysiology. Global fluctuations in blood-oxygenation-level-dependent (BOLD) fMRI signals appear across various networks following sleep deprivation (7, 8), and large-scale fMRI signals are linked to altered electrophysiology, eyelid closures (9), and arousal state, suggesting global changes in brain activity and vigilance (10–13). Attentional lapses after sleep deprivation are linked to reduced activity within thalamus and cognitive control areas, suggesting a failure to engage large-scale network activity (14–16). At the scale of local cortical areas, attentional failures are linked to sleep-like low-frequency waves in isolated cortical patches in rats (17). Intriguingly, transient increases in low-frequency activity also predict attentional lapses in humans (18, 19). Together, these findings have shown that attentional deficits following sleep deprivation are linked to brainwide hemodynamic changes and low-frequency neural oscillations. A key open question is what drives the brain to generate these spontaneous drops in neural arousal state and behavior after sleep deprivation.

One possibility is that attentional lapses correspond to a brief state in which the brain transiently carries out a sleep-dependent function that is incompatible with waking behavior. Sleep serves many functional purposes, and several studies have shown that one critical role is to clear the neurotoxic waste products that accumulate during wakefulness (20–24); although a recent study reported the opposite pattern (25). Importantly, the conflicting studies used different sleep protocols for the awake measurements (rested wakefulness vs. sleep-deprived), raising the question of whether sleep deprivation could alter this process. Waste clearance in the brain is mediated by cerebrospinal fluid (CSF), and CSF begins to pulse in large, low-frequency (0.01–0.1 Hz) waves during non-rapid eye movement sleep (NREM) (26, 27). In addition, sleep deprivation modulates pulsatility and solute concentration in the CSF (28, 29), suggesting the effects of sleep deprivation could be accompanied by altered CSF dynamics. However, whether behavioral deficits are linked to CSF flow has not been explored.

We conducted a within-subject total sleep deprivation study in healthy human participants to investigate whether sleep deprivation and its associated attentional deficits are linked to altered brain fluid dynamics. We used multimodal fast fMRI, EEG, behavioral assessments, and pupillometry, to track multiple aspects of neurophysiological dynamics simultaneously. We discovered that CSF flow is coupled to behavioral deficits during wakefulness after sleep deprivation. Notably, we found that attentional failures are marked by a brain and bodywide state change that underlies both behavioral deficits and pulsatile fluid flow, and that the moments where attention fails signal the occurrence of a large-scale pulse of CSF flow in the brain.

Results:

To investigate how sleep deprivation modulates neural activity and CSF dynamics, we performed a simultaneous EEG and fast fMRI experiment in 26 human participants. To enable within-subject comparisons, each participant was scanned twice: once after a night of regular sleep (well-rested), and once after one night of total sleep deprivation, which was continuously supervised in the laboratory (Fig. 1A). We performed simultaneous EEG-fMRI scans with pupillometry in the morning. During the scans, subjects performed up to four runs of a sustained attention task, the psychomotor vigilance test (PVT, Fig. 1B), followed by an eyes-closed resting-state run.

Fig. 1. After sleep deprivation, CSF flow exhibits large sleep-like low-frequency waves during wakefulness.

(A) Sleep-deprived (SD) visit: Subjects arrived at the laboratory at 7PM prior to the sleep-deprived night, and were monitored continuously during the night. Well-rested (WR) visit: Subjects arrived between 8:30–9am the day of the scan. For both visits, scans were performed around 10am in the morning. Scans included up to four PVT runs followed by a 25-minute resting-state run. (B) The PVT attention task: Each run used either the auditory PVT (detecting a beep) or the visual PVT (detecting a luminance-matched visual stimulus). (C) Left: Example fMRI acquisition volume position (green box) for simultaneous measurement of BOLD and CSF flow. The volume intersects the fourth ventricle, to enable upwards CSF flow detection. Yellow marks the cortical segmentation; purple indicates flow measurement ROI. Image masked to delete identifiers. Right: Example placement of CSF ROI (magenta) in the fourth ventricle in one representative subject. (D) CSF timeseries from the same subject during wakefulness shows that sleep deprivation causes large CSF waves during wakefulness, whereas the well-rested (WR) condition shows smaller CSF flow. (E) Sleep deprivation increased low frequency (0.01–0.1 Hz) CSF power during wakefulness (Awake n=486 segments in WR; 205 SD; N1 n=179 WR; n=59 SD; N2 n=57 WR; n=40 SD), black bar indicates p<0.05, permutation test with Bonferroni correction), to a magnitude similar to N1 and N2 sleep. Power spectral density (PSD) calculated on CSF signal in resting-state runs. Shading is standard error. (F) Paired analysis of CSF low-frequency power in resting-state wakefulness shows increased power after SD (n=18 subjects with artifact-free wakefulness at both sessions, paired t-test, Bonferroni corrected).

Sleep deprivation causes sleep-like pulsatile CSF flow to intrude into wakefulness

We first investigated whether sleep deprivation altered the dynamics of CSF flow in resting-state scans that included awake and sleeping periods (Fig. S1). Our fast fMRI protocol enabled us to simultaneously measure cortical hemodynamic BOLD signals and CSF inflow in the fourth ventricle (Fig. 1C). Consistent with previous studies, the CSF inflow signal during well-rested wakefulness exhibited a small-amplitude rhythm synchronized to respiration (30, 31), in contrast to NREM sleep where CSF exhibits large ~0.05 Hz waves (26). However, visual inspection clearly showed that the CSF signal during wakefulness after sleep deprivation also exhibited large-amplitude low-frequency waves, resembling NREM sleep (Fig. 1D). We analyzed this CSF signal across sleep stages in participants who had at least 60 seconds of continuous wakefulness during the resting-state run (n=22), excluding segments with high motion (framewise displacement>0.5mm), and found a 4.7 dB increase in the CSF signal peaking at 0.04 Hz in sleep-deprived wakefulness (Fig. 1E). Remarkably, the CSF power in sleep-deprived wakefulness reached levels similar to the CSF power during typical N2 sleep (0.01–0.1 Hz power in SD-wake: −14.96 dB, 95%CI=[−15.88, −14.04]); rested N2: −15.13 dB, 95%CI=[−16.37, −13.88]). Hemodynamics are a key driver of CSF flow, as changes in blood vessels can mechanically drive CSF flow (26, 32–34), and we found that sleep deprivation also induced a significant increase in low-frequency (0.01–0.1 Hz) cortical gray matter BOLD power (p=0.0048, paired t-test). These session-dependent changes in CSF and BOLD were not driven by motion artifacts (Fig. S2). These results demonstrated that sleep deprivation caused low-frequency CSF flow pulsations and hemodynamic waves to appear during wakefulness, resembling the dynamics typically seen during N1 or N2 sleep.

Pulsatile CSF flow occurs during epochs with worse attentional task performance

Since large-scale CSF waves typically occur during NREM sleep, a state in which attention is suppressed, we investigated whether the CSF waves during sleep-deprived wakefulness were associated with any attentional cost. As expected (2, 14, 35–41), sleep deprivation caused an increase in the mean reaction time (Fig. 2A), and the omission (missed response) rate (Fig. 2B) in the PVT task. Surprisingly, we observed that attentional lapses (RT>500ms) and omissions tended to coincide with higher-amplitude CSF flow (Fig. 2C). To quantify this behavior-CSF relationship, we tested whether lapses (RT>500ms) and omissions were linked to CSF power, analyzing data in 60-s segments during confirmed wakefulness (EEG and no eyelid closures>1s). We found that worse attentional performance was linked to increased low-frequency (0.01–0.1 Hz) CSF power: segments with attentional lapses and segments with omissions exhibited significantly higher CSF flow power than segments without lapses (Fig. 2D). This effect remained significant when controlling for motion (two-way ANOVA; p>0.05 for motion). These results demonstrated that larger low-frequency CSF flow is associated with attentional failures during wakefulness.

Fig. 2. Pulsatile CSF flow dynamics increase during epochs with slower reaction times and attentional failures.

(A) Reaction times (RTs) after sleep deprivation showed higher mean and a longer tail, indicating more behavioral lapses (Mann-Whitney U test, p<0.001). (B) Omission rate increased after sleep deprivation (n=26 subjects, p<0.001, paired t-test). (C) CSF flow and reaction time fluctuations during one example run, showing higher flow when reaction time slows down. (D) Low-frequency (0.01–0.1 Hz) CSF power within non-overlapping 60s segments, categorized into three different states: high attention (all RTs below 500 ms); low attention (at least one RT>500 ms), and omissions (at least one omission). Higher CSF power appears in lower attentional states (p<0.001 for main effect; one-way repeated measures ANOVA and Tukey post hoc test, n=26 subjects).

What process could link CSF flow to attentional function? Ascending neuromodulation, such as noradrenergic input from the locus coeruleus, modulates attentional function (42, 43), and also act directly on the vasculature (44), making this a candidate mechanism for modulating both attention and CSF flow. Neuromodulatory tone and locus coeruleus firing is correlated with pupil diameter (45), so we investigated whether pupil diameter was linked to CSF flow on faster timescales. We found that pupil diameter was significantly correlated with CSF flow: constricted pupil was linked to downwards CSF signals, and dilated pupil was linked to subsequent upwards CSF signals (Fig. 3 A&B). This correlation remained significant when we tested segments from rested wakefulness and after sleep deprivation (Fig. S3). Consistent with this, we also observed that CSF flow peaks were locked to drops in attentional performance: both reaction time and the rate of omissions increased before CSF peaks (Fig. 3 C&D). In each case, while CSF flow was locked to behavior and pupil, it lagged these measures in time, which could reflect a vascular mechanism which would introduce a time delay. Furthermore, while pupil diameter is consistently correlated with noradrenergic tone, it is also correlated with several other modulators (46, 47). To test whether pupil-CSF coupling reflected a pupil-locked modulation of cerebrovascular fluctuations, which could be expected from noradrenergic-driven vasoconstriction and would drive a delayed CSF response (48), we also examined the global cortical BOLD signal, and found an anticorrelation with the pupil (Fig. 3E). We calculated the best-fit impulse response linking pupil to the global BOLD signal, and found a negative impulse response with a peak delay of 6.75s, consistent with pupil-linked vasoconstriction. To test whether this mechanism in turn predicted CSF flow, we convolved the pupil signal with its impulse response and calculated the expected CSF flow driven by this vascular effect, with no additional parameter fitting (Fultz et al., 2019). We found that the convolved pupil signal yielded a significant prediction of CSF activity (Fig. 3H, zero-lag R=0.26, maximal R=0.3 at lag −1.75s), indicating that the timing of the pupil-CSF coupling could be explained by a vascular intermediary.

Fig. 3. Pulsatile CSF flow is temporally coupled to pupil diameter changes and behavioral performance during wakefulness.

(A) Spontaneous pupil constriction and dilation is time locked to peaks of CSF flow. Black bars indicate significant changes in z-scored pupil diameter compared to baseline (p<0.05, t-test, baseline = [−30 −28] s, Bonferroni corrected). (B) Cross correlation between pupil diameter and CSF showed strong correlation (maximal r = 0.26 at lag −4.25s; n = 709 segments, 26 participants) (C) During period with pupil constriction, reaction times also showed significant increase compared to same baseline (p<0.05, t-test, Bonferroni corrected). (D) Omission rate during task showed significant increase during pupil constriction, and significant decrease during pupil dilation (p<0.05, t-test, Bonferroni corrected). (E) Significant biphasic changes in cortical BOLD activity are locked to CSF peaks (p<0.05, t-test, Bonferroni corrected). (F) Mean CSF signal. (G) To estimate the impulse response function linking pupil size changes to BOLD and CSF activity, we convolved pupil diameter traces from each segments with a series of impulse response function (IRF). Estimated impulse response of the cortical BOLD signal to the pupil diameter shows a time-to-peak at 6.75s. Cross-correlation between pupil diameter and BOLD showed strong correlation (maximal r = 0.38 at lag 0s; n = 709 segments, 26 participants). (H) Predicting CSF flow with no additional parameter fitting, assuming that the derivative of the pupil-locked BOLD fluctuations drives CSF flow, shows significant prediction of the true CSF signals (zero lag r = 0.26; n=709 segments).

CSF pulsatile flow is temporally coupled to attentional failures and altered brain state

Having found that CSF flow was highest in epochs with omissions, we next examined the precise dynamics occurring at the moment of omissions. Since CSF flow increases during NREM sleep, we carefully excluded omissions due to NREM sleep, to confirm whether these flow changes appeared during wakefulness (EEG-verified, and excluding segments with eye closures>1s). We found a striking coupling between behavior and CSF flow during sleep-deprived wakefulness: omissions were locked to a downward wave and then an upward wave of CSF flow in the fourth ventricle (Fig. 4A). While few omissions occurred in the well-rested state, when they did appear they showed similar CSF flow coupling, demonstrating that a complete attentional failure even when rested can engage a similar mechanism (Fig. S4A). The cortical gray matter BOLD signal exhibited a matched biphasic pattern, consistent with vascular drive of downwards and then upwards CSF flow (Fig. S4B). However, our imaging protocol was only able to directly measure CSF inflow (upwards to the brain) due to the placement of the acquisition volume (Fig. 1C), whereas drops in CSF signal could in theory represent either outflow or simply no flow. We therefore conducted a second study in an additional 10 sleep-restricted subjects with a new imaging protocol that measured bidirectional flow (Fig. S6A) to test whether this pattern specifically signaled CSF outflow before inflow. We confirmed that omissions were locked to a pulse of CSF flowing downward out of the brain, beginning at the time of the missed stimulus, followed by a pulse flowing upward into the brain (Fig. S6B). These results replicated the finding of coupled attentional function and CSF flow in an independent cohort, with a biphasic profile of downwards flow at omissions, followed by upwards flow.

Fig. 4. Attentional failures are coupled to pulsatile CSF flow and a series of neural and systemic physiological changes.

(A) Awake omission trials are locked to a biphasic change in CSF flow, with a downwards trough followed by an upwards peak. Time zero marks stimulus onset; omission occurs within the following seconds. (B) Omission trials are locked to a broadband drop in EEG power, including decreased alpha-beta (10–25Hz) EEG power then an increase in slow wave activity (SWA; 0.5–4 Hz) and alpha-beta power. The biphasic alpha-beta power change is widespread with centro-occipital predominance, whereas the SWA increase has occipital and frontal predominance. Spectrogram is normalized within-frequency; black bars indicate significant a change in broadband power (0.5–30Hz) compared to baseline (p<0.05, t-test, baseline=[−20 −10] s, Bonferroni corrected). (C) The aperiodic component of the EEG was subtracted to display oscillation-specific changes. The alpha-beta and SWA effects were still present (statistics in Fig. S5). (D) Pupil diameter showed a biphasic change at omissions, with a significant constriction during the omission trial followed by dilation after trial onset (9.00–12.00s). (E) Heart rate dropped during omissions and subsequently increased. (F) Respiratory rate dropped at omissions and subsequently increased. Black bars with stars indicate significant changes from baseline (n=364 trials, 26 subjects, p<0.05, paired t-test, Bonferroni corrected). Shading is standard error.

During NREM sleep, CSF flow is coupled to neural slow wave activity (21, 25), and local slow waves can also increase after sleep deprivation (17). We therefore analyzed the EEG spectrogram to investigate whether neural dynamics might be linked to these omission-locked CSF waves. We found that omissions (n=364 trials, 26 subjects) were accompanied by EEG alterations across multiple frequency bands, with a drop in broadband EEG power, particularly pronounced in the alpha-beta (10–25Hz) range, followed shortly by subsequent increase in power (Fig. 4B). In addition to this broadband change, the oscillation-only component of the spectrogram (calculated by subtracting the aperiodic component (49–52) also demonstrated significant clusters at slow and alpha-beta frequencies, with a broad spatial distribution (Fig. 4C, Fig S5). Overall, this broadband power reduction and steepening spectral slope pointed to a spatially distributed change in electrophysiological dynamics.

What type of neuronal event might these EEG spectral shifts reflect? The relatively broadband nature of these EEG changes suggested that they might reflect a transient change in cortical excitability (53), perhaps mediated by central neuromodulatory state, as suggested by the pupil coupling (Fig. 3). In addition, autonomic state changes and systemic oscillations can drive CSF flow (54–56), consistent with the idea that large-scale neuromodulatory changes could contribute to both the attentional failures and CSF dynamics. We therefore tested whether a sequence of events could underlie the CSF flow: a drop in attentional state followed by recover. Consistent with this, pupil dilation was biphasically-coupled to CSF flow: the pupil was constricted during the omission, corresponding to a low arousal state and downward CSF flow, and then subsequently dilated (Fig. 4D), with CSF then flowing upward (Fig. 4A). Since neuromodulatory state also has systemic effects (57), we further examined systemic physiological recordings, including changes in respiratory rate and volume, heart rate, and peripheral vascular volume – reflecting activity of the autonomic nervous system. We found that all systemic measures were significantly locked to the omission (Fig. 4D–F, Fig. S7). These results demonstrate that CSF inflow events during wakefulness were coupled to a brain-and-bodywide integrated state shift, manifesting as a transient behavioral deficit, electrophysiological markers of distinct neuronal state, and pupil constriction.

CSF pulsatile flow is differentially modulated by loss and recovery of attention

These results clearly identified widespread behavioral, neuronal, and arousal dynamics that were locked to CSF pulsatile flow during sleep-deprived wakefulness. However, attentional failures are typically brief during wakefulness, with a spike in arousal rapidly following an omission. Due to this biphasic pattern, this CSF-locked result could be explained by two different possibilities: either that CSF flow was specifically linked to drops in arousal (at the onset of attentional failures) or that it was locked to increases in arousal (at recovery of behavior). We therefore designed an analysis of behavioral responses to separate the neurophysiological dynamics linked to the drop, vs. recovery, of attention. We categorized behavioral omissions into three types: “Type A”, where the omission trial was preceded and followed by a valid response; “Type B”, the first of at least three consecutive omissions; and “Type C”, the last omission of a consecutive series (Fig. 5A). As above, wakefulness during all omissions was verified both through EEG scoring and eye-tracking.

Fig. 5. CSF flows outwards when attention drops, and back in when attention recovers.

(A) Omissions during wakefulness were categorized as isolated (Type A); onset of sustained low attention (Type B); and end of sustained low attention (Type C). Schematic made with BioRender. (B) Type A, isolated omission trials (n=127 trials, 26 subjects) were locked to a decrease in broadband EEG power, then increased power at attention recovery in the next trial. CSF showed a significant decrease followed by a significant increase. Respiratory rate and heart rate both showed significant changes from baseline after trial onset. Orange arrow points to the average timing (8.1s) of the first valid response after the isolated omission. (C) Type B: first omission of the series signifies the onset of a sustained low attentional state (n=61 trials, 20 subjects). CSF signal, respiratory rate and heart rate all decrease significantly (p<0.05, paired t-test, Bonferroni corrected). Orange arrow points to the average timing (25.1s) of the first valid response after the series of omissions. (D) Type C: last omission of the series, followed by increased attentional state (n=57 trials, 20 subjects). EEG shows heightened SWA during the omission and subsequent increase in alpha-beta power at recovery of attention. CSF signal, respiratory rate and heart rate all increase significantly (p<0.05, paired t-test, Bonferroni corrected). Orange arrow points to the average timing (9.0s after time 0) of the first valid response after the series of omission. For all panels in B-D: Black bars indicate significant (p<0.05, paired t-test, Bonferroni corrected) changes from baseline ([−10 −5] s). Shading is standard error.

Remarkably, this analysis identified two separable patterns: neural, CSF and systemic signals showed opposing changes locked to Type B (arousal drops) and C (arousal increases) omissions (Fig. 5B, C, D). The loss of attentional focus in Type B omissions was linked to a decrease in EEG alpha-beta power and expulsion of CSF, whereas regaining of attentional focus in Type C omissions co-occurred with an increase in EEG broadband power and upwards flow of CSF, and the systemic arousal indicators followed the same pattern (Fig 5C, D). These results demonstrated that CSF flow occurs in a specific sequence of events: during an attentional failure, EEG power dropped in concert with bodywide signatures of low arousal, and a pulse of CSF flows downwards out of the brain. This pattern inverted at recovery of arousal, with CSF returning upwards as attention improved, with a time delay consistent with vascular-mediated CSF flow. These results identified widespread behavioral, neuronal, and arousal dynamics that were locked to CSF pulsatile flow during sleep-deprived wakefulness, consistent with an integrated neuromodulatory system governing neural arousal state and CSF flow.

Discussion:

Here, we found that attentional failures after sleep deprivation are locked to a brain- and body-wide state change and pulsatile waves of CSF flow. During sleep-deprived wakefulness, CSF flow pulsations intruded into the awake state, causing a flow pattern that resembled N2 NREM sleep. Strikingly, CSF waves were strongly coupled to behavioral failures, a shift in neuronal spectra, and pupil constriction. Specifically, attention dropped when CSF flowed outward, and attention recovered as CSF was drawn back upwards into the brain. The attentional failures that occur after sleep deprivation thus reflect the initiation of a process in which CSF is transiently flushed out of the brain.

Our data showed a consistent pattern of broadband EEG shifts linked to attentional failures. What type of neural event might these represent? The process of falling asleep is linked to suppression of EEG alpha power and drops in respiratory rate (58, 59), which could suggest that the events we observe here represent an accelerated transition toward sleep onset. In addition, autonomic arousal fluctuations are characterized by a sequential modulation of beta power and slow wave activity (54, 60, 61). The EEG pattern we observe here shares similar broadband features, and we find that it represents a behaviorally relevant modulation of neural vigilance state, with attentional consequences. This pattern may thus reflect that sleep deprivation causes sleep-initiating cortical events to occur during wakefulness, but with an interruption of the initiation process before sleep is attained. Given that we found a shift in spectral slope, this pattern could also be consistent with increasing cortical inhibition (62). While we could not directly identify local sleep slow waves here due to the macroscopic nature of these measurements, this effect appeared to be relatively spatially widespread, as it was detected across many EEG electrodes.

The neural and attentional drops were also linked to pupil constriction, a classic signature of arousal state that is actively controlled by noradrenergic neuromodulatory projections (63, 64). Pupil diameter is correlated with attention, heart rate, and galvanic skin reflex, reflecting a tight coupling between the state of the central and peripheral arousal systems (12, 47, 65). Furthermore, the autonomic system can strongly modulate CSF flow, with strong effects during light sleep (55). The high correlation between pupil diameter and CSF pulsatile flow thus suggests the involvement of the ascending neuromodulatory system in regulating CSF flow during wakefulness (66–68). A possible mechanistic explanation for the joint changes in EEG, CSF, and autonomic indicators is the well-established parallel projections from key neuromodulatory nuclei such as the locus coeruleus, which projects throughout cortex and thalamus to modulate neural state, while also directly acting on the sympathetic nervous system (69–72). Intriguingly, locus coeruleus activity oscillates during NREM sleep and drives infraslow neuronal spectral changes (73, 74), and pupil diameter also covaries with these neural signatures (75, 76); our results suggest a similar dynamic of temporally structured noradrenergic fluctuations could appear during sleep-deprived wakefulness. Other neuromodulators are also correlated with pupil diameter and could contribute (46, 77), although our hemodynamic data suggest a substantial role for the vasoconstrictive effects of the noradrenergic system (78). All of the dynamics we report were observed during eyes-open wakefulness, demonstrating that brief shifts in attentional state are sufficient to elicit these coordinated changes. Our results could thus be consistent with sleep deprivation causing a destabilization of the ascending arousal system, with brief failures of its wakefulness-promoting actions leading to attentional dysfunction, neuronal arousal suppression, and CSF flow.

A key question is why these attentional shifts produce such large-scale CSF flow, detectable even in the ventricle. The neuromodulatory mechanism could potentially explain these dual effects. A primary driver of CSF flow is changes in the vasculature, as dilation and constriction of blood vessels in turn propels CSF flow (26, 32, 33, 79–83). Since noradrenaline is a vasoconstrictor (44), drops in locus coeruleus activity could decrease neuronal arousal state while dilating cortical blood vessels and constricting the pupil. This would result in a downwards flow in the fourth ventricle detected several seconds later, which then would invert as attention recovers and noradrenergic activity increases, reflected in the subsequent pupil dilation. An exciting implication of these dynamics is that pupil diameter could potentially provide a noninvasive, accessible readout of signals related to CSF flow.

These findings may impact the interpretation of previous fMRI experiments. While common practice is often to regress out CSF signals to attempt to remove noise, our data show that this regression-based approach will also remove or distort attention-related signals of neural origin that are collinear with the CSF signals – which are often signals of interest when studying brain dynamics. Moreover, CSF flow signals reported here are within eyes-open wakefulness, suggesting that even during the awake state, fluctuations in attention could have strong effects on CSF signals. Notably, CSF flow signals linked to eye closure have been detected in the widely used Human Connectome Project dataset (84), suggesting that regressing out CSF may influence a large set of fMRI studies.

A surprising finding of this study is that attentional function is coupled to CSF flow, with slower reaction times and attentional failures corresponding to the initiation of CSF flow out of the brain. Prolonged wakefulness leads to buildup of toxic metabolic waste products (22, 23, 29, 85), suggesting that more fluid flow would be needed after sleep deprivation. Why would behavioral performance and fluid dynamics then be tightly coupled at rapid timescales within the sleep-deprived awake state? One possibility is that they are not directly functionally related, but rather that the same circuit controls both and thus they covary. In this case, drops in ascending arousal could drive both behavioral changes (to decrease waste production) and fluid changes (to increase waste transport). The timing of events we observed could align with this hypothesis, since the downward CSF flow in the fourth ventricle began at the onset of the missed stimulus, and could reflect a change in the cortex and subarachnoid space seconds earlier. An alternate possibility is that prolonged CSF flow is incompatible with stable attentional states, for example because high flow rates alter the concentration of signaling molecules that are required for typical waking function. We found that while CSF flow was tightly coupled to behavior, it did not exclusively occur during omissions – CSF flow pulses continuously with lower amplitude during wakefulness (30, 56, 86), and flow increased to a lesser degree during successful performance with slow reaction times (Fig. 2, 3) – meaning that a single large event of CSF flow does not prevent behavior. However, it is possible that sustained high-amplitude flow over longer timescales interferes with behavioral performance, leading to a structured alternation between high-attention and high-flow states.

Overall, our results demonstrate that the attentional failures caused by sleep deprivation are coupled to large-scale brain fluid transport. When the brain lacks the opportunity to sleep, it enters a sub-optimal attentional state, which may provide partial benefits of sleep at the cost of behavioral errors. This coupling points to a central circuit that controls both attentional state and CSF flow, and shows that the moments of attentional dysfunction we experience after sleep loss correspond to the emergence of widespread fluid flow in the brain.