Night terrors: the screaming sleep episode your child will not remember
Last updated September 3, 2026.
Night terrors are partial-arousal episodes from deep sleep: a child (typically 3-8 years) suddenly screams, thrashes, sweats, and stares with wide eyes, inconsolable and not truly awake, for minutes, then returns to sleep with no memory in the morning. They look terrifying and are almost always harmless: a developmental quirk of the child's sleep system, strongly linked to overtiredness, and outgrown by the teen years in the large majority.
What does an episode look like?
The signature: 1-4 hours after falling asleep (the deep-sleep window), the child sits up or bolts screaming, eyes open but unseeing, heart pounding, sweating, maybe muttering or thrashing, and cannot be comforted or woken properly (trying to wake them prolongs it). It lasts a few minutes to (rarely) half an hour, then ends as abruptly, and the child sleeps on with total amnesia: only the parents carry the memory. Nightmares, the main contrast, happen in the second half of the night, wake the child fully, and are remembered.
Why do they happen?
The child's brain partially wakes from the deepest sleep stage, stuck between sleep and waking: it runs in families (sleepwalking and night terrors share the trait), and the triggers are consistent: overtiredness (the biggest), irregular sleep schedules, illness and fever, a full bladder, a noisy sleep environment, stress, and some medications. They are not caused by trauma, bad dreams, or bad parenting, and they say nothing about the child's psychological health. Peak age is 3-8 years; most children outgrow them by adolescence.
What actually helps?
- During an episode: do not try to wake the child (it prolongs and intensifies); stay close, keep them safe from falls and knocks, speak calmly, and let it pass.
- Fix the overtiredness: the highest-yield move: an earlier bedtime, a consistent sleep schedule (weekends too), and enough total sleep for the age.
- Scheduled awakening: if episodes come at a predictable time: gently rouse the child (not fully) 15-30 minutes before the usual episode time for a few weeks: it breaks the cycle for many.
- Reduce the triggers: a toilet trip before bed, a cool quiet room, and treating any fever or illness.
- Safety-proof: especially if sleepwalking joins in: stair gates, locked windows and doors, nothing hazardous on the floor.
When is it an emergency?
Night terrors almost never need urgent care. See the GP for: episodes very frequent or worsening, episodes starting in the second half of the night with stiffening, jerking, or tongue-biting (seizures are the important mimic: a video of an episode helps the doctor enormously), loud snoring or breathing pauses (sleep apnea drives night terrors and is treatable), daytime sleepiness or behavior change, injury risk, or episodes persisting into the teens. Pymander's escalation routing is built and tested specifically for this class of decision; see the safety architecture working paper.
What a Pymander AI doctor consult looks like
Illustrative example, not a real member's messages.
Common questions
Should I wake my child during a night terror?
No: trying to wake them is the one well-meaning move that backfires. The child is in deep sleep throughout (the open eyes deceive), forced waking prolongs and intensifies the episode, and a half-woken terrified child is harder to settle than one left to ride it out. The right response is supervision: stay nearby, move them gently away from edges and hard objects, speak calmly and quietly (for you as much as them), and wait: episodes end on their own in minutes, and the child sleeps on peacefully. The morning rule completes it: do not interrogate or alarm them about it; they remember nothing, and making it a thing creates bedtime anxiety where none existed.
Are night terrors a sign of trauma or emotional problems?
No: night terrors are a maturation quirk of the sleep system (the brain catching between deep sleep and waking), not a window into distress: the research is consistent that they reflect overtiredness, genetics, and triggers like fever and full bladders, not psychological trauma, and children with them are as emotionally healthy as any other. (Nightmares, the remembered kind, do spike with stress and anxiety: a different phenomenon, second half of the night, fully waking.) The parental instinct that something terrible must have happened is understandable and unfounded. The emotional burden of night terrors lands on the parents, who watch them, not the child, who sleeps through them.
What is the scheduled awakening technique?
The cleverest low-tech intervention in sleep medicine: if episodes arrive at a predictable time (they usually do, 1-4 hours into sleep), you gently rouse your child 15-30 minutes before the habitual episode time, just enough to stir them (a light touch, a murmured word, a half-sit-up: not full waking), then let them resettle, repeated nightly for two to four weeks. The rousing interrupts the deep-sleep pattern the terror was riding, and success rates in studies are high for habitual-time episodes. Keep a brief sleep diary first to time it accurately. It feels strange to wake a sleeping child; doing it deliberately is exactly what stops the involuntary version.
Could it be seizures? How would we know?
The distinction parents need: night terrors come 1-4 hours after sleep onset (deep-sleep window), involve screaming, thrashing, and unresponsiveness with morning amnesia, and the child is floppy-normal between episodes; seizures in sleep more often cluster in the lighter sleep of the second half of the night, involve stereotyped stiffening, rhythmic jerking, or tongue-biting, may come in brief clusters, and can leave the child confused or headachey. When the description is ambiguous, the diagnostic tool is already in your pocket: video an episode on your phone, which lets the doctor sort it in seconds. Any stiffening, rhythmic jerking, or daytime episodes earn a prompt GP visit.
Will he grow out of it, and when?
Almost certainly yes: night terrors peak between 3 and 8 years and the large majority of children outgrow them by adolescence as the deep-sleep architecture matures (the same maturation that ends most sleepwalking). The family pattern (a parent who sleepwalked or had night terrors) predicts the tendency but not the duration. The frequency often fades over months once overtiredness is managed: the earlier, consistent bedtime is doing more than it looks like. Persistence into the teen years or adult onset is the point to seek a sleep assessment, because adult-onset episodes deserve a proper look at triggers, sleep apnea, and rarer causes.
Does snoring or sleep apnea matter here?
Yes, and it is the treatable connection worth checking: obstructive sleep apnea in children (usually big tonsils and adenoids) fragments deep sleep and is a documented driver of night terrors and sleepwalking, with episodes often resolving after the apnea is treated. The screening questions: loud habitual snoring, witnessed breathing pauses, gasping or choking sounds, restless sleep with odd positions (neck extended), mouth breathing, and daytime crankiness or inattention (children go hyper rather than sleepy). If any fit, tell the GP: the tonsil-and-adenoid assessment is straightforward, and fixing the airway frequently fixes the nights for the whole house.