There is a category of advice that feels intuitively right but turns out, when examined carefully, to be wrong in the direction you expected. Evening wind-down routines are not that category. The science here is unusually convergent: the practices that improve sleep are well-identified, the mechanisms are understood, and the evidence for several specific interventions is strong enough to have influenced clinical guidelines. What is less understood, and what popular accounts often miss, is the why -- the specific biological mechanisms that make evening behavior so consequential for sleep quality.
Sleep is not simply the absence of wakefulness. It is an active, highly organized biological process that the body begins preparing for hours in advance. By the time you lie down, your core body temperature should already be dropping, melatonin should already be rising, adenosine should have accumulated to levels that make sleep pressure irresistible, and the cognitive hyperarousal that keeps people awake should be winding down. If the two hours before bed have involved bright screens, stressful news, a heated argument, vigorous exercise, or a glass of wine, the body's preparation for sleep has been disrupted in measurable ways -- and the sleep that follows will be measurably worse, even if you fall asleep quickly.
What follows is an evidence-based account of what the evening transition to sleep actually involves biologically, and which practices are supported by research. The goal is not a prescriptive ritual but an understanding of the mechanisms that allows you to construct one that fits your life and biology.
"Sleep is the single most effective thing we can do to reset our brain and body health each day." -- Matthew Walker
Key Definitions
Adenosine: A chemical byproduct of neuronal metabolic activity that accumulates in the brain and cerebrospinal fluid throughout the waking day, building sleep pressure by binding to receptors that promote drowsiness and suppress wakefulness. Adenosine is cleared during sleep, which is why a full night of sleep produces the refreshed, low-sleep-pressure state of the following morning.
Sleep pressure: The drive to sleep, driven primarily by adenosine accumulation, that increases with time awake and decreases during sleep. Sleep pressure combines with the circadian rhythm (which regulates the timing of wakefulness and sleep) to determine when sleep onset occurs and how deep sleep is.
Melatonin: A hormone produced by the pineal gland that signals to the body's circadian clock that it is time for sleep. Melatonin does not induce sleep directly but shifts physiological systems toward sleep readiness. Its production is suppressed by light, particularly short-wavelength blue light.
Cognitive hyperarousal: A state of elevated mental activity, rumination, worry, and problem-solving that is a primary driver of sleep-onset difficulty and chronic insomnia. Cognitive hyperarousal involves activation of the default mode network and maintains wakefulness even when physiological sleep pressure is high.
CBT-I: Cognitive Behavioral Therapy for Insomnia. A structured psychological treatment targeting the thoughts and behaviors that perpetuate insomnia, including hyperarousal, maladaptive beliefs about sleep, and conditioned wakefulness. Currently the recommended first-line treatment for chronic insomnia.
Adenosine and Sleep Pressure: The Chemical Case for Wind-Down
How Sleep Pressure Builds
Every neuron in your brain metabolizes ATP (adenosine triphosphate) for energy. The byproduct of this process includes adenosine, which is released into the extracellular space and cerebrospinal fluid. As waking hours accumulate, adenosine levels rise, progressively binding to receptors that promote drowsiness and inhibit the arousal circuits that maintain wakefulness. This is what sleep scientists mean by "sleep pressure" -- a chemical pressure that builds across the day and is only relieved by sleep.
The sleep pressure system works in concert with the circadian clock (which regulates the timing of wakefulness and sleep regardless of sleep pressure) to produce the characteristic pattern of human alertness: relatively high in the morning, dipping in the early afternoon, rising again in the late afternoon, and then declining in the evening as both sleep pressure and the circadian timing toward sleep converge.
Caffeine works entirely by blocking adenosine receptors -- it does not reduce adenosine accumulation, it temporarily prevents the signal from reaching its target. This is why caffeine consumed in the afternoon or evening (with a half-life of approximately 5 to 7 hours) delays sleep onset and reduces sleep quality even when people feel they can fall asleep normally: the adenosine is still there, waiting for the caffeine to clear. Evening caffeine is among the most common and most underappreciated disruptors of sleep quality.
Blue Light and Melatonin: Czeisler's Research
The Harvard Studies
Charles Czeisler at Harvard Medical School and the Division of Sleep and Circadian Disorders at Brigham and Women's Hospital has led some of the most influential research on light's effects on human circadian biology and sleep. His group's work established that light is the dominant zeitgeber (time-giver) for the human circadian clock, and that short-wavelength light in the 480-nanometer blue range is the most potent suppressor of melatonin production.
A 2014 study by Anne-Marie Chang and colleagues in Czeisler's lab, published in the Proceedings of the National Academy of Sciences, compared reading on a light-emitting tablet before bed to reading a printed book under dim light. Participants in the tablet condition showed delayed melatonin onset (approximately 90 minutes), took longer to fall asleep, showed altered sleep stage architecture, reported reduced morning alertness, and showed suppressed morning melatonin -- effects that persisted for several days after the study. The tablet readers also showed reduced subjective sleepiness at bedtime, meaning the device was interfering with the biological signal of sleep readiness without participants necessarily noticing.
The practical implication is not that all screens must be avoided entirely in the evening, but that the intensity, duration, and timing of blue-light-emitting screen use meaningfully affects the biological preparation for sleep. Screen use in the 1 to 2 hours before bed, particularly at high brightness, consistently produces measurable shifts in sleep timing and quality.
Practical Mitigation
Blue light filtering software (f.lux, Night Shift, Night Mode) shifts screen color temperature toward the amber range in the evening, reducing blue light output. Evidence suggests this is a partial mitigation -- it reduces but does not eliminate the melatonin-suppressing effect. Screen brightness reduction is complementary: lower intensity reduces stimulation even without color temperature adjustment. Physical blue-light-blocking glasses have mixed evidence but may provide additional benefit for heavy evening screen users.
The most effective intervention remains reducing screen time in the 60 to 90 minutes before bed, regardless of filtering. Reading physical books, gentle stretching, low-light conversation, or other screen-free activities in the pre-sleep window eliminates the problem at its source.
CBT-I: The Evidence-Based Treatment for Insomnia
What CBT-I Addresses
Jack Edinger at the National Jewish Health in Denver and others have been central to the development and validation of Cognitive Behavioral Therapy for Insomnia (CBT-I). CBT-I is a structured, typically 6 to 8-session treatment that targets the psychological and behavioral factors that perpetuate chronic insomnia, which are often more significant than the factors that originally triggered it.
The core CBT-I components include: sleep restriction therapy (temporarily limiting time in bed to match actual sleep time, building sleep pressure and improving sleep efficiency); stimulus control (using the bed only for sleep and sex, getting out of bed when awake for more than 20 minutes, to break the conditioned association between bed and wakefulness); sleep hygiene education (addressing behaviors that disrupt sleep); relaxation training (reducing physiological and cognitive arousal); and cognitive restructuring (challenging catastrophic beliefs about the consequences of poor sleep that themselves maintain hyperarousal).
Multiple meta-analyses have found CBT-I to be as effective as sleep medication (particularly benzodiazepines and Z-drugs) in the short term, and significantly more effective in the long term. Medication effects disappear when the medication is discontinued; CBT-I effects persist and often improve after treatment ends, because they address the perpetuating factors rather than merely suppressing symptoms. The American College of Physicians, the American Academy of Sleep Medicine, and equivalent organizations in multiple countries now recommend CBT-I as the first-line treatment for chronic insomnia, ahead of pharmacotherapy.
Scullin's To-Do List Study (2018)
Writing Upcoming Tasks Promotes Sleep Onset
A 2018 study by Michael Scullin and colleagues at Baylor University used polysomnography -- objective, electrode-based measurement of sleep architecture -- to test a simple question: does writing before bed help people fall asleep faster? Participants were randomly assigned to spend five minutes writing either a to-do list of upcoming tasks they needed to complete in the coming days, or a list of tasks they had already completed.
The to-do list writers fell asleep significantly faster -- about 9 minutes faster on average. The more specific and comprehensive the to-do list, the faster the sleep onset. The completed-tasks writers showed no such benefit.
The proposed mechanism is cognitive offloading: the persistent mental activity of rehearsing uncompleted tasks (the Zeigarnik effect -- the tendency for incomplete tasks to remain in active memory) is a recognized driver of pre-sleep rumination. By writing down the upcoming tasks in detail, participants effectively transferred the responsibility for remembering them to the external record, which reduced the mind's need to keep rehearsing them. The thoroughness effect (more specific lists led to faster sleep) is consistent with this interpretation: a vague or partial capture does not fully offload the rehearsal demand.
This is a simple, free, low-risk intervention that is consistent with the broader CBT-I literature on cognitive hyperarousal as a driver of insomnia.
Hot Bath Timing: Haghayegh et al. (2019)
The Thermoregulation Mechanism
Shahab Haghayegh and colleagues at the University of Texas at Austin published a systematic review and meta-analysis in 2019 examining the effects of pre-sleep bathing on sleep quality. The review analyzed 17 studies involving both warm baths and warm showers taken before bed and examined outcomes including sleep onset latency, sleep quality ratings, and objective sleep measures.
The analysis found that bathing in water between 40 and 42.5 degrees Celsius (104 to 108.5 degrees Fahrenheit), taken 1 to 2 hours before bedtime, significantly improved sleep onset (average reduction of approximately 10 minutes), subjective sleep quality, and several objective sleep measures. The timing window of 1 to 2 hours before bed was critical: bathing immediately before bed was less effective.
The mechanism is thermoregulatory. Sleep onset is associated with a drop in core body temperature of approximately 1 to 2 degrees Celsius. The body achieves this by increasing blood flow to the extremities (hands and feet), which dissipates heat from the core. A hot bath accelerates this process: it raises skin temperature and dilates peripheral blood vessels, increasing blood flow to the hands and feet, which then radiates heat to the environment. After leaving the bath, core temperature drops more rapidly than it would through natural evening cooling alone. This accelerated temperature drop mimics and amplifies the biological signal that triggers sleep onset.
Alcohol and Sleep: The REM Disruption Problem
Why Alcohol Is Not a Sleep Aid
Alcohol is a central nervous system depressant with sedative effects that reduce sleep onset time. This creates the widespread belief that it helps sleep. The physiological reality is more nuanced and largely negative.
Alcohol at typical evening consumption levels (one to three drinks) does reduce the time to fall asleep. However, as alcohol is metabolized (typically 4 to 5 hours after consumption), it produces a second-half-of-night disruption: increased wakefulness, lighter sleep stages, suppressed REM sleep, and more frequent awakenings. REM sleep is the stage most important for emotional memory processing, creative problem-solving, and the emotional regulation work of sleep. Research by Matthew Walker and others has shown that even moderate alcohol consumption substantially suppresses REM sleep, reducing its proportion of total sleep time regardless of whether total sleep duration appears normal.
The physiological consequence is sleep that looks adequate by duration but is impaired in architecture. People who drink regularly in the evenings and report sleeping fine often show the REM suppression and second-half fragmentation on polysomnography that is not captured by subjective reports. The tolerance effect -- the feeling that you sleep fine with alcohol because you have habituated to the disrupted pattern -- masks the real cost.
The 90-Minute Wind-Down Window
Why Transition Time Matters
The transition from active wakefulness to sleep-ready physiology takes time. Physiological systems that support wakefulness (elevated core temperature, cortisol, sympathetic nervous tone, alertness-promoting neurotransmitters) do not switch off instantly. The circadian and adenosine systems begin shifting toward sleep in the evening, but external inputs -- bright light, vigorous exercise, stressful cognitive demands, stimulating social interaction, alcohol, caffeine -- can delay or disrupt this transition.
A 90-minute wind-down window provides the biological preparation time that sleep onset requires. This is not an arbitrary number: it reflects the approximate time needed for the combination of core body temperature decline, melatonin rise, adenosine signaling, and cortisol reduction to progress to levels that support smooth sleep onset. Research on sleep hygiene, CBT-I, and circadian interventions converges on this approximate window as the meaningful threshold for pre-sleep behavior change.
The specific content of the wind-down window matters less than its general character: lower light, lower stimulation, lower cognitive demand, and the reduction of blue-light screen use. Activities that fit this profile include reading physical books, gentle stretching or yoga, low-intensity conversational interaction, journaling (including the to-do list approach of Scullin's research), and the hot bath described by Haghayegh's review.
Cognitive Shuffling: Beaudoin's Sleep-Onset Technique
Interrupting the Rumination Loop
Luc Beaudoin at Simon Fraser University developed a sleep-onset technique he called cognitive shuffling, based on a theoretical model of the brain's transition from wakefulness to sleep. Beaudoin proposed that the brain uses a process of "serial diverse imagining" to disengage from coherent, purposeful thought and enter the hypnagogic state that precedes sleep. The brain signals that it is safe to transition to sleep partly by generating low-significance, unconnected imagery.
In people with sleep-onset difficulties, this transition is often blocked by coherent, sequential thought: worry about tomorrow, planning, reviewing the day, or narrative thinking. These high-coherence thought patterns maintain the cognitive vigilance that keeps the brain awake.
Cognitive shuffling interrupts this pattern by deliberately generating a stream of random, unconnected mental images: a banana, then a lighthouse, then a bicycle, then a giraffe, without any narrative connection between them. The randomness prevents the brain from engaging in the sequential problem-solving that maintains wakefulness and provides a stream of low-significance content that mimics the hypnagogic imagery the brain naturally produces as it drifts toward sleep. The technique is taught in several apps, including MySleepButton. Preliminary controlled studies have supported its effectiveness for sleep-onset difficulties.
Practical Takeaways
Begin the wind-down 90 minutes before intended sleep time. This is enough time for core temperature to begin dropping, melatonin to rise, and cortisol to reduce to sleep-promoting levels. Compression of this window consistently delays or fragments sleep.
Reduce screen brightness and blue light exposure. Use night mode, reduce brightness, or switch to screen-free activities in the 60 to 90 minutes before bed. The melatonin suppression from bright screens is real and measurable.
Write a specific to-do list for tomorrow before bed. Based on Scullin's 2018 research, detailed capture of upcoming tasks offloads the rehearsal demand that drives pre-sleep rumination. Invest five to ten minutes; be specific rather than vague.
Take a warm bath or shower 1 to 2 hours before bed. The thermoregulatory mechanism described by Haghayegh's review is well-supported. Timing matters: too close to bed does not allow core temperature to drop before sleep onset.
Treat alcohol as a sleep disruptor, not a sleep aid. The sedative effect is real; the sleep improvement effect is not. If sleep quality is a priority, evening alcohol consumption is among the most impactful variables to address.
Avoid vigorous exercise within 2 to 3 hours of bed. Vigorous exercise elevates core temperature, heart rate, and sympathetic nervous tone in ways that conflict with the physiological preparation for sleep. Morning or early afternoon exercise avoids this conflict entirely and offers focus benefits as described in how to improve focus.
If you cannot sleep, do not lie in bed awake. Stimulus control (a core CBT-I component) prevents the conditioned association between the bed and wakefulness. Get up, go to a dim room, do a quiet activity, and return to bed when sleepy. One week of consistent application produces measurable improvement in sleep efficiency. The morning routine follows from the night before -- see what makes a good morning routine for how consistent wake timing anchors the whole cycle.
References
- Chang, A. M., Aeschbach, D., Duffy, J. F., and Czeisler, C. A. "Evening Use of Light-Emitting eReaders Negatively Affects Sleep, Circadian Timing, and Next-Morning Alertness." Proceedings of the National Academy of Sciences, 2015.
- Scullin, M. K., Krueger, M. L., Ballard, H. K., Pruett, N., and Bliwise, D. L. "The Effects of Bedtime Writing on Difficulty Falling Asleep: A Polysomnographic Study Comparing To-Do Lists and Completed Activity Lists." Experimental Brain Research, 2018.
- Haghayegh, S., Khoshnevis, S., Smolensky, M. H., Diller, K. R., and Castriotta, R. J. "Before-Bedtime Passive Body Heating by Warm Shower or Bath to Improve Sleep: A Systematic Review and Meta-Analysis." Sleep Medicine Reviews, 2019.
- Edinger, J. D., and Carney, C. E. Overcoming Insomnia: A Cognitive-Behavioral Therapy Approach. Oxford University Press, 2008.
- Walker, M. Why We Sleep: Unlocking the Power of Sleep and Dreams. Scribner, 2017.
- Czeisler, C. A., and Gooley, J. J. "Sleep and Circadian Rhythms in Humans." Cold Spring Harbor Symposia on Quantitative Biology, 2007.
- Borbely, A. A. "A Two Process Model of Sleep Regulation." Human Neurobiology, 1982.
- Morin, C. M., and Espie, C. A. Insomnia: A Clinical Guide to Assessment and Treatment. Springer, 2004.
- Beaudoin, L. P., and Digdon, N. "Bedtime Ruminative Thinking and Cognitive Shuffling as Interventions for Sleep-Onset Problems." SLEEP, 2019.
- Trotti, L. M. "Waking Up Is the Hardest Thing I Do All Day: Sleep Inertia and Sleep Drunkenness." Sleep Medicine Reviews, 2017.
- Roehrs, T., and Roth, T. "Sleep, Sleepiness, and Alcohol Use." Alcohol Research and Health, 2001.
- Epstein, L. J., and Mardon, S. The Harvard Medical School Guide to a Good Night's Sleep. McGraw-Hill, 2006.
Article Word Count: 2,842
Frequently Asked Questions
What is adenosine and how does it create sleep pressure?
Adenosine is a chemical byproduct of neuronal activity that accumulates in the brain throughout the waking day. As adenosine builds up, it binds to receptors that promote drowsiness and suppress wakefulness, creating what sleep scientists call 'sleep pressure' or 'sleep drive.' The longer you have been awake, the more adenosine has accumulated and the stronger the drive to sleep. Caffeine works by blocking adenosine receptors, which is why it temporarily suppresses sleepiness without reducing the underlying accumulation: when caffeine clears, the adenosine is still there. During sleep, adenosine is cleared from the brain, which is why a full night of sleep produces the refreshed feeling of low sleep pressure the following morning.
How does blue light affect melatonin and sleep?
Research led by Charles Czeisler at Harvard Medical School and colleagues has demonstrated that light, particularly short-wavelength blue light (around 480 nanometers), is the most potent suppressor of melatonin production by the pineal gland. Melatonin is the hormone that signals the body's internal clock that it is time to sleep: it does not induce sleep directly but shifts the circadian phase toward sleep onset. Evening exposure to screens (phones, tablets, computers, televisions) emitting blue-enriched light delays melatonin onset, postpones sleep timing, and reduces sleep duration. Czeisler's group estimated that a two-hour exposure to a bright tablet before bed delayed melatonin onset by about 90 minutes compared to reading a printed book.
What is CBT-I and why is it recommended over sleep medication?
Cognitive Behavioral Therapy for Insomnia (CBT-I) is a structured psychological treatment developed and validated by researchers including Jack Edinger. It addresses the thoughts and behaviors that perpetuate insomnia: hyperarousal, unrealistic beliefs about sleep, unhelpful coping strategies like extended time in bed, and conditioned wakefulness in the bedroom. Multiple meta-analyses have found CBT-I to be as effective as sleep medication in the short term and more effective in the long term, with effects that persist after treatment ends while medication effects disappear when discontinued. Major medical organizations including the American College of Physicians recommend CBT-I as the first-line treatment for chronic insomnia. Its core components include sleep restriction therapy, stimulus control, and sleep hygiene education.
Does writing a to-do list before bed actually help sleep?
A 2018 study by Michael Scullin and colleagues at Baylor University used polysomnography (objective sleep measurement) to test whether writing a to-do list before bed helped participants fall asleep faster. Participants who spent five minutes writing a detailed to-do list of upcoming tasks they needed to remember fell asleep significantly faster than those who wrote about completed tasks. The researchers proposed that offloading upcoming tasks onto paper reduces the cognitive burden of mentally rehearsing them, which is a common cause of pre-sleep rumination. The more specific and detailed the to-do list, the faster participants fell asleep, suggesting that thoroughness matters: partial offloading does not eliminate the rehearsal loop as effectively as complete capture.
Why does a hot bath help sleep and when should you take it?
A 2019 systematic review by Shahab Haghayegh and colleagues analyzed 17 studies and found that a warm bath or shower taken 1 to 2 hours before bedtime, at a temperature between 40 and 42.5 degrees Celsius, accelerated sleep onset by an average of about 10 minutes and improved subjective sleep quality. The mechanism involves thermoregulation: core body temperature needs to drop by approximately 1 to 2 degrees to initiate sleep. A hot bath raises skin temperature, which increases blood flow to the hands and feet, which in turn dissipates heat from the body's core. The result is a faster-than-normal drop in core temperature once you leave the bath, which mimics and accelerates the natural cooling signal the brain uses to trigger sleep onset. Timing matters: too close to bed (less than 30 minutes) does not allow enough cooling time.
Does alcohol help or hurt sleep?
Alcohol is a sedative that does reduce sleep onset time, which creates the misleading impression that it improves sleep. However, as alcohol is metabolized during the night (typically 4 to 5 hours after consumption), it produces a 'rebound' effect that fragments the second half of sleep and suppresses REM sleep. REM sleep is the stage most important for emotional memory consolidation, creative problem-solving, and mood regulation. Research consistently shows that even moderate alcohol consumption before bed reduces REM sleep duration, increases nighttime awakenings, and reduces overall sleep quality even when total duration appears normal. The sedating effect of alcohol is real; the sleep-improving effect is not.
What is cognitive shuffling and how does it help you fall asleep?
Cognitive shuffling is a technique developed by cognitive scientist Luc Beaudoin at Simon Fraser University, based on a model of how the brain transitions from wakefulness to sleep. Beaudoin proposed that the brain uses a 'serial diverse imagining' process to reduce cognitive coherence and signal that safety and wakefulness are no longer needed. The technique involves deliberately generating a loose, random sequence of unconnected mental images: imagining a banana, then a lighthouse, then a bicycle, then a mountain, without narrative connection. This interrupts the sequential, problem-solving thought patterns that keep people awake by providing a stream of low-significance content that mimics the hypnagogic imagery the brain produces naturally as it drifts toward sleep. Preliminary studies have supported its effectiveness for sleep-onset difficulties.