The brains of people with insomnia disorder show a reduced ability to resist auditory disturbances during sleep, particularly in deep sleep, according to a study published in iScience. The work introduces the concept of “sleep resilience” as a measurable trait and suggests that objective EEG markers of neural instability could help characterize insomnia beyond subjective symptom reports.
Researchers at Beijing Normal University and collaborating institutions recruited 58 patients with insomnia disorder and 43 healthy controls for overnight sleep recordings with event-locked auditory stimulation. They measured how the brain responded to brief sounds presented during sleep, using a framework that distinguished three dimensions of resilience: resistance (the ability to prevent neural activation), containment (the ability to limit the spread of activation once triggered), and recovery (the ability to return to baseline).
What they found
Compared with healthy controls, insomnia patients showed:
- Reduced resistance: Evoked brain responses (measured by EEG complexity and P300 amplitude) were significantly larger in insomnia patients, especially during N3 (deep) sleep. This indicates that the sleeping insomnia brain is more easily perturbed by external stimuli.
- Impaired containment: Insomnia patients exhibited stronger fronto-parietal phase-amplitude coupling during the middle post-stimulus window, suggesting that once a disturbance triggered brain activation, the neural circuits were less effective at limiting its spread.
- Relatively preserved recovery: In the late post-stimulus window, the abnormalities in the insomnia group weakened, consistent with an ability to eventually return to baseline — though the initial vulnerability remained.
The findings were most pronounced in N3 sleep, the deepest stage of NREM sleep, which is normally the most stable and resistant to disruption. This suggests a specific deficit in the neural mechanisms that normally protect deep sleep from environmental perturbations.
Why it matters
Insomnia is conventionally diagnosed and tracked through subjective reports — difficulty falling asleep, staying asleep, or waking too early. Objective markers of sleep quality have been harder to come by. This study identifies EEG-based measures of “sleep resilience” that could serve as biomarkers for insomnia severity and treatment response.
The finding that the deficit is most severe in deep sleep is noteworthy, because N3 sleep is when the brain typically shows the highest threshold for arousal. A breakdown of this protection may help explain why people with insomnia often describe their sleep as “light” or “unrefreshing” even when total sleep time appears normal on polysomnography.
Limits
The sample is moderate (58 patients, 43 controls), and the auditory stimulation paradigm, while well-controlled, represents a laboratory setting that may not fully capture the types of naturally occurring sleep disruptions people experience at home. The study measured responses to external sounds, not endogenous arousals, which may have different neural signatures. Longitudinal data tracking whether these markers change with treatment would strengthen the case for their clinical use.
Bottom line
Insomnia disorder is characterized by impaired neural resilience to disturbance during deep sleep, measurable through EEG markers of evoked brain activity. This framework could provide objective, physiology-based endpoints for insomnia research and treatment evaluation.
Source: Yang C, Gu H, Chen H, et al. Impaired sleep resilience underlies transient neural instability in insomnia disorder. iScience. 2026 Jun 19;29(6):116151. DOI: 10.1016/j.isci.2026.116151. PMID: 42305591. Open access.