Set a restricted sleep window equal to your actual sleep time
Limit time in bed to approximately your average actual sleep time — this is the restriction that starts the cure.
Key takeaways
- What it is: Limit time in bed to approximately your average actual sleep time — this is the restriction that starts the cure.
- Why it works: Restricting time in bed forces consolidated sleep by ensuring that homeostatic pressure (Process S) is high before the window opens and maintained throughout. The shorter window prevents the fragmented, spread-thin sleep that low-efficiency insomnia produces. As the brain learns that the bed means sleep (not prolonged wakefulness), the conditioned arousal that maintains insomnia is gradually extinguished.
- Evidence: Backed by randomized trials / meta-analyses.
- Avoid: Setting the window to how long you want to sleep rather than how long you are actually sleeping — which fails to build the pressure the protocol depends on. The window must match reality, not aspiration.
Why it works
Restricting time in bed forces consolidated sleep by ensuring that homeostatic pressure (Process S) is high before the window opens and maintained throughout. The shorter window prevents the fragmented, spread-thin sleep that low-efficiency insomnia produces. As the brain learns that the bed means sleep (not prolonged wakefulness), the conditioned arousal that maintains insomnia is gradually extinguished.
How to do it
- 1Take your average total sleep time from the baseline log and set your sleep window to that duration, with a minimum of five and a half hours.
- 2Choose a consistent wake time first, then calculate your bedtime by subtracting the window duration.
- 3Do not get into bed before your designated bedtime, regardless of how tired you feel.
What the evidence says
RCT / meta-analysisSleep restriction is one of the best-supported single components of CBT-I; it improves sleep efficiency faster than sleep hygiene alone and has durable effects.
Honest caveat: Sleep restriction increases daytime sleepiness during the first one to two weeks and can pose safety risks for people who drive or operate heavy machinery; it should not be used without awareness of this. Contraindicated in people with bipolar disorder, seizure history, or parasomnias.
- — Spielman, Saskin & Thorpy (1987), treatment of chronic insomnia by restriction of time in bed, Sleep
- — Morin et al. (2006), psychological and pharmacological treatments for insomnia, JAMA
Common mistake
Setting the window to how long you want to sleep rather than how long you are actually sleeping — which fails to build the pressure the protocol depends on. The window must match reality, not aspiration.
IX Coach calculates your starting sleep window from the sleep log and adjusts your target bedtime to the minute, removing the decision from a point in the evening when you may be tired and tempted to cheat.
Practice this with IX Coach →