CBT without a Therapist
Start free
GUIDE · SLEEP

Paradoxical intention for insomnia: gently trying to stay awake

You've been in bed so long that trying to fall asleep has become the main activity. The harder you aim at sleep, the more awake you feel. Paradoxical intention changes the instruction: lie in the dark and gently try to stay awake. It's an old technique with a small, mixed research base. This page covers where it comes from, how it's meant to work, how to try it carefully, and where the evidence runs out.

Reviewed by our editors~8 min read

What paradoxical intention is and where it comes from

Paradoxical intention for insomnia means getting into bed and gently trying to stay awake instead of trying to fall asleep. You keep the lights off and your eyes open, and you drop the job of making sleep happen. Our guide to CBT for insomnia mentions it briefly, and this page goes deeper.

The idea comes from Viktor Frankl, the Viennese psychiatrist who developed logotherapy. The psychiatrist Benoît Bayle writes that Frankl began working it out in 1929, described it in a scientific article in 1939, and named it in 1947. For anxiety and obsessions, Frankl had people wish for the very thing they feared. The Viktor Frankl Institute says the aim is to use self-distancing and irony to break the loop in which a symptom reinforces itself.

A 1978 report by Ascher and Efran described five people with trouble falling asleep who were told to try to remain awake. A 2023 review says the patients reported difficulty following the instruction because they fell asleep too quickly. A 1979 paper by Ascher and Turner gave the wording the 2023 review calls the most common: lie comfortably in bed in a dark room with your eyes open and try to stay awake as long as possible, without turning to activities that are incompatible with sleep.

Why removing the effort might help

The original explanation is about effort. Ascher and Efran proposed that trying to fall asleep creates frustration and arousal: you monitor yourself, feel anxious about performing, try harder, and fail again. Paradoxical intention is meant to lower that performance anxiety so natural sleep processes can take over. A later model by Espie and colleagues makes a similar argument. Sleep runs fairly automatically, so focused attention and direct attempts to control it can inhibit it.

One experiment tested the idea. Broomfield and Espie (2003) followed 34 people with sleep-onset insomnia through a seven-night baseline, then randomly assigned them to 14 nights of paradoxical intention or to a control group. Compared with controls, the paradoxical intention group showed a significant drop in sleep effort and sleep performance anxiety. An activity monitor showed no difference in time to fall asleep, though self-reported time trended lower. The authors suggested the monitor may be insensitive, or that people with insomnia overestimate how long they lie awake.

The experiment fits the theory's prediction about effort and anxiety without showing that measured sleep improved. Other proposed routes, such as exposure to being awake in bed or a buildup of sleep pressure, remain hypotheses.

Who it may suit, and who it may not

No study has shown who benefits. The 2023 review says there's no research on matching people to the technique, so its suggestions rest on theory. The review's obvious candidate struggles to fall asleep and acts as if sleep can and should be controlled. According to the review, seven of the 10 trials in a 2022 meta-analysis enrolled people with trouble falling asleep and three included a mix, so there's little to go on for waking up at night.

Three of the five trials in the American Academy of Sleep Medicine (AASM) review excluded people with medical conditions, so we know little about how it works when another health condition is involved. The AASM review also warns that it may raise anxiety about sleep in some people and may appeal less to people who already struggle to fall asleep. Bayle lists suicidal thoughts and post-traumatic stress as reasons not to use Frankl's original method. That article isn't about sleep, but it's a reason to see a clinician first if lying awake brings up painful memories or hopelessness.

Skip it the night before driving a long distance, running machinery, or doing other work where drowsiness is dangerous. The studies we found didn't measure next-day sleepiness.

A worked example (invented)

This example is invented to show the shape of a night. It isn't data, and your nights will differ.

Dana has spent months willing herself to sleep, checking the clock and counting the hours left. She decides to try paradoxical intention for two weeks. She goes to bed when she's sleepy, turns the lights off, leaves her phone in the kitchen, and turns the alarm clock to the wall. She tells herself: "I'll lie here with my eyes open. I don't have to fall asleep."

On the first night she catches herself picturing a calm beach and lets it go, because that's a sleep trick. Some nights she drifts off mid-instruction and notices only in the morning. On another night her heart pounds with frustration, so she gets up, reads a paper book in a dim room, and returns when she's sleepy.

When it backfires

Secretly trying to fall asleep: If you stay awake hoping to trick yourself into sleep, the effort is still running under a new label. Falling asleep partway through is fine.

Adding sleep tricks: Counting breaths, relaxing your muscles in sequence, or playing cognitive shuffling are all attempts to bring sleep on. They bring back the effort this technique is meant to remove.

Watching the clock: In two experiments by Tang, Schmidt and Harvey (2007), people told to monitor a clock while trying to sleep reported more pre-sleep worry. In the first experiment they also took longer to fall asleep than people who didn't watch a clock.

Expecting quick results: Early nights can go the other way. The 2023 review reports that in a 1989 trial a third of participants had longer sleep-onset times during the first week of paradoxical intention, though times had shortened by the end of treatment. In a case series of six people, sleep onset got longer for three. Give the trial its full length, and stop sooner if it leaves you more anxious.

How it fits with stimulus control and CBT-I

The AASM guideline strongly recommends multicomponent CBT-I for chronic insomnia in adults. The 2023 review sees paradoxical intention as a possible second option after CBT-I has been considered or tried. It also warns that the technique clashes with stimulus control, which tells you to get up when you can't sleep, and with sleep restriction, which shrinks your time in bed. The authors advise using these one after the other instead of on the same nights.

What the research does and doesn't say

The research is small, mostly old, and mixed. A 2022 meta-analysis of 10 trials by Jansson-Fröjmark and colleagues found large improvements in key insomnia symptoms against passive comparators, and smaller but moderate ones on several central outcomes against active comparators. The authors called for methodologically stronger studies.

The AASM's 2021 systematic review was less hopeful. It found five randomized trials and could judge clinical significance in only two. One found sleep latency 28.25 minutes lower with paradoxical intention than with control (95% CI 8.37 to 48.13 minutes). The other found 1.38 minutes lower (95% CI from 19.50 minutes lower to 16.74 minutes higher), below the clinical significance threshold. The review rated the evidence very low quality because of imprecision, inconsistency, and risk of bias, and said harms were mostly unstudied.

The AASM's 2021 clinical practice guideline makes no recommendation about paradoxical intention, because fewer than three studies met its inclusion criteria. It adds that paradoxical intention and the other therapies it couldn't rate can still be considered when CBT-I is unavailable, unwanted, or unsuitable. Earlier AASM papers were kinder: the 2006 practice parameters listed paradoxical intention among individually effective therapies at the Guideline level.

As far as we could find, no trial has tested what paradoxical intention adds to a full CBT-I program. The 2023 review says studies of daytime effects, side effects, adherence, and long-term follow-up are missing. A 2024 protocol for a 40-person uncontrolled pilot notes that no study has run in the last few decades, and we didn't find published results. Treat it as a technique worth a careful trial. The evidence doesn't support calling it a proven fix.

A two-week trial of paradoxical intention

The studies we read give no per-night time limit beyond staying awake as long as possible. Broomfield and Espie used 14 nights, so this plan uses two weeks. Read the cautions above first.

  1. Choose a fair two-week stretch. Pick two weeks with no long drives or hazardous work on a short night. If you're working on stimulus control, finish or pause it first, since the two pull in opposite directions.
  2. Write the instruction in your own words. Something like: I'll lie here with my eyes open and stay awake as long as I comfortably can. I don't have to fall asleep. Leave out every sleep trick, including counting, breathing patterns, and word games.
  3. Go to bed when you're sleepy. Use your normal bedtime, turn the lights off, put your phone out of reach, and turn any clock away from you.
  4. Lie with your eyes open, softly. Rest your gaze on the dark without straining. Gently repeat your instruction whenever you catch yourself checking how close to sleep you are. If sleep arrives, let it.
  5. Keep an exit rule for bad nights. If you're awake and frustrated, or your heart is pounding, get up. Sit somewhere dim, read on paper, and return when you're sleepy. This comes from stimulus control, not from the paradoxical intention trials, so treat it as a safety valve.
  6. Jot one line each morning. Write how the night felt, without minutes or sleep scores, which invite the same checking as a clock. After 14 nights, reread your notes. If nothing changed or you feel more anxious, stop and look at CBT for insomnia or talk to a clinician.

Expect a few rougher nights early. The short Sleep Test asks how often sleep problems have bothered you over the past month, and anxiety at night covers the worry side of bedtime.

Plan your own trial with Helpy

The chat below runs with Helpy, the site's AI assistant. Describe what your bedtimes look like, and it can help you word the instruction in your own voice and think through what a fair two-week trial would involve.

Evidence and sources

When lying awake needs a clinician's eyes

If sleep trouble has lasted a few months, you snore loudly or stop breathing at night, or sleepiness is affecting your driving or work, talk with a doctor or a sleep specialist. The same goes for severe anxiety, a low mood that won't lift, or painful memories that surface when you lie awake. Staying awake on purpose isn't the right experiment then. If you're thinking about hurting yourself, call or text 988 or call 911.

Important

This is educational self-help content, not a substitute for professional care. If anxiety is intense, sticks around for weeks, or makes daily life hard, reach out to a mental-health professional or doctor. If you’re in crisis or thinking about suicide, get help now: call or text 988 (988 Suicide & Crisis Lifeline), text HOME to 741741 (Crisis Text Line), or call 911 for emergencies. Available 24/7.

Read next

How to beat insomnia without pills: CBT-I, step by stepThe full program this technique belongs toCognitive shuffling: the falling-asleep word game, done rightAnother way to quiet a racing mind in bedAnxiety before sleepWhy your thoughts race at night

← All CBT guides