Lead
We have all known someone who seems immune to sleep loss. They pull an all-nighter, show up the next morning, and still function. Then there is the rest of us who, after one bad night, struggle to keep our eyes open, let alone think clearly. Why do some people handle sleep deprivation better than others? A new study published today in Sleep points to a surprising answer: it may come down to how stable certain brain networks remain over the course of prolonged wakefulness.
Researchers at the University of Arizona College of Medicine and the University of Michigan put 16 healthy adults through a grueling 39-hour total sleep deprivation protocol and scanned their brains six times across the first 32 hours. They found that individuals who maintained their cognitive performance despite mounting sleep pressure had strikingly more stable functional connectivity in two specific brain subnetworks, one linking the thalamus with cortical regions, and another connecting visual and memory-related areas.
What they found
The study, led by Jungwon Cha and senior author William D.S. Killgore at the University of Arizona’s Social, Cognitive, and Affective Neuroscience (SCAN) Lab, used resting-state functional MRI to track changes in brain connectivity as sleep deprivation set in. Participants were kept awake in the lab under continuous supervision, with no naps, caffeine, or stimulants allowed. At regular intervals, they climbed into the scanner for a resting-state scan and then completed the Psychomotor Vigilance Task (PVT), a gold-standard test of sustained attention that measures reaction times to a visual stimulus.
By the end of the protocol, performance on the PVT had diverged sharply. Some participants maintained relatively stable reaction times throughout the 39 hours. Others showed progressive slowing, the classic signature of a brain losing its battle with sleep pressure.
When the researchers analyzed the fMRI data using network-based statistics, two subnetworks emerged as key predictors of who would hold up and who would fall apart.
The first was a thalamocortical subnetwork involving the thalamus, the globus pallidus, and widespread cortical regions. The thalamus acts as the brain’s sensory relay station, gating information flow to the cortex. Its connections became progressively weaker over time in low-resilience individuals, while high-resilience participants maintained stable connectivity across the same circuit.
The second was a perceptual-memory subnetwork linking visual cortex with memory-related areas. Again, the pattern was clear: stable connectivity meant stable performance. Those whose visual-memory connections held steady over the course of sleep deprivation continued to respond quickly on the PVT. Those whose connectivity declined saw their reaction times rise accordingly.
Importantly, the effect was not about who started out with stronger connections. It was about trajectory. High-resilience individuals did not necessarily have more connectivity at baseline. What distinguished them was that their connectivity did not deteriorate over time. The brain networks that supported attention and alertness remained resilient even as the body clock ticked past its normal wake period.
Why it matters
Sleep loss is pervasive in modern life. Shift workers, medical residents, military personnel, long-haul truck drivers, and new parents all face extended periods of wakefulness, often when critical decisions are on the line. Understanding why some individuals are more vulnerable to sleep deprivation, and why others seem protected, has real-world implications for safety, performance, and health.
The findings suggest that stable functional connectivity in thalamocortical and perceptual-memory networks may serve as a neural signature of resilience to sleep loss. If these patterns can be detected in real time, it might be possible to identify when a person is approaching cognitive failure before they are aware of it themselves.
The researchers note that wearable EEG or even advanced fMRI-based monitoring could one day provide fatigue-state warnings. Imagine a system that alerts a tired surgeon, truck driver, or air traffic controller that their brain networks are beginning to destabilize and it is time for a break. That kind of real-time fatigue monitoring could prevent accidents caused by sleep-deprived errors.
The study also adds to a growing body of evidence that resilience to sleep loss is not a single trait but reflects the stability of specific neural circuits. This reframes the question from “Are you resilient to sleep deprivation?” to “Which of your brain networks are resilient?” and points toward targeted interventions that might shore up vulnerable circuits before they fail.
Limits
The study is small, 16 participants, and the findings will need to be replicated in larger, more diverse samples. The 39-hour total sleep deprivation protocol, while tightly controlled, does not perfectly mirror real-world sleep loss, which is often partial, chronic, or intermittent rather than one continuous stretch of wakefulness. Resting-state fMRI captures brain activity when participants are not engaged in any task, so how these connectivity patterns translate to real-world decision-making under fatigue remains an open question. The researchers also note that resilience was defined using a single behavioral measure (PVT reaction time), and broader cognitive assessments might reveal different patterns.
Bottom line
Stability, not strength, is what matters. People who maintain cognitive performance during sleep deprivation do not necessarily have stronger brain connections at the start. They have connections that do not weaken over time. The thalamocortical and perceptual-memory subnetworks identified in this study may act as a neural buffer against the effects of prolonged wakefulness, and monitoring their stability could one day help predict, and prevent, fatigue-related failures in high-stakes environments.
Source
Cha J, Negelspach D, Huskey A, Kennedy K, Katz J, Forger D, Killgore WDS. Functional brain network stability reflects individual resilience during sleep deprivation. Sleep. 2026 Jul 21. DOI: 10.1093/sleep/zsag201. PMID: 42477887.