Usually, the conversation within psychiatry eventually gets to medication. Should we add trazodone? A Z-drug? Low-dose doxepin? Melatonin? Something else? And, of course, the gold standard for chronic insomnia remains cognitive behavioral therapy for insomnia, or CBT-I.

But a 2026 study published in Proceedings of the National Academy of Sciences made me think about sleep from a slightly different direction. Instead of asking how we get someone to sleep, the researchers asked a more practical question: what can we do about the brain after someone has already lost the sleep?

One of the research answers was surprisingly simple. Short Duration Exercise.

The 30-Hour Problem.

First, an important distinction. This was not an insomnia study.

The researchers studied healthy young adults, ages 18 to 35, who were kept awake for 30 hours. They then compared what happened when participants did one of three things: a 20-minute bout of moderate-to-vigorous cycling, a 90-minute nap, or essentially nothing (Lotlikar et al., PNAS, 2026).

Thirty hours awake sounds extreme. Unfortunately, it isn't completely foreign to the world many of our patients live in. Think about the client working a double shift, the parent caring for a sick child, someone working two jobs while also caring for an aging parent, or the nurse working nights because that schedule is the only way the family can make everything work.

We have more tools designed to make life easier than ever before, yet somehow many people seem busier than ever. Sometimes sleep doesn't just become disrupted. It gets lost. That is the population I had in mind while reading this paper.

After 30 hours without sleep, both interventions helped. A single 20-minute bout of moderate-to-vigorous cycling and a 90-minute nap protected the participants' ability to form new episodic memories. On average across the two interventions, recognition performance was roughly 22% better than in the inactive control group—about 21% for exercise and 23% for the nap (Lotlikar et al., PNAS, 2026).

Even more interesting was that exercise and napping appeared to get there through different pathways. The nap reduced EEG markers associated with sleep pressure and neural fatigue. That makes intuitive sense: the brain was sleep deprived, so giving it some sleep helped. Exercise was different. It didn't appear to simply erase the accumulated sleep pressure; instead, the EEG findings suggested the brain became more efficient at processing and encoding information afterward.

☕️ ☕️ In other words, exercise didn't replace sleep. It appeared to help the sleep-deprived brain work better despite the lack of sleep.

I'll add one caution here. The EEG analyses involved only about 13 to 15 people per group, the key brain–behavior correlations were within-group rather than formal between-group comparisons, and one of the headline statistical models improved substantially after a single influential data point was removed. So, I read the mechanism as an interesting hypothesis, that will need more exploration.

And there was another finding I found particularly interesting: exercise didn't make participants more fatigued. The exercise group reported no greater fatigue than the group that essentially did nothing, yet they performed better on the memory task.

Before We Start Prescribing Pelotons

There are some substantial limitations here.

The study was small with 18 participants per group for the behavioral analysis, with even fewer for portions of the EEG analysis. It was a single laboratory session.

More importantly, these were healthy young adults experiencing acute total sleep deprivation. They were not 72-year-olds with depression, chronic pain, and six medications. They were not patients with generalized anxiety disorder who have been sleeping four fragmented hours a night for 9 months. And they were not people with chronic insomnia.

Chronic insomnia involves different physiology, including hyperarousal, conditioned associations with the bed and sleep, and sometimes a mismatch between perceived and objectively measured sleep.

But there is a much more interesting clinical question hiding underneath it.

Exercise May Be One of Our Underused Prescriptions

Exercise already has considerably stronger evidence in chronic insomnia than this one study. Moderate aerobic exercise has been associated with improvements in sleep-onset latency, wake after sleep onset, total sleep time, sleep efficiency, and overall sleep-quality scores, with several plausible mechanisms: circadian regulation, HPA-axis effects, BDNF, inflammatory signaling, and changes in slow-wave sleep (Saadh et al., Sleep Medicine, 2025).

For our patients specifically, the psychiatric evidence is worth knowing. A 2026 systematic review and meta-analysis found that exercise significantly improves sleep quality across psychiatric disorders—with some of the largest effects in depression, anxiety, and schizophrenia—and identified a practical "dose": low-to-moderate intensity, at least three sessions per week, at least 30 minutes per session, in programs shorter than 12 weeks (Wang et al., BMC Psychiatry, 2026). That gives us something concrete to prescribe rather than a vague suggestion.

And now this newer research adds another piece: even a single bout of exercise may help protect daytime cognitive function after substantial sleep loss.

That matters because when patients complain about insomnia, they aren't always complaining about the night. They're complaining about the next day. "I can't think." "I can't concentrate." "I'm exhausted at work." "My memory is terrible." Those complaints often drive treatment-seeking just as much as lying awake at 2 a.m.

So perhaps part of our conversation shouldn't only be, How do we make you sleep tonight? It may also be, How do we protect tomorrow?

This is where we need to be careful. Someone could look at this study and conclude that if a 90-minute nap helped memory, perhaps we should recommend naps to patients with insomnia. Not so fast.

Standard CBT-I generally discourages daytime napping, and there is a physiologic reason. One of the things we're trying to build throughout the day is homeostatic sleep pressure, which is the increasing biological drive to sleep that accumulates the longer we stay awake. A substantial daytime nap can partially discharge that pressure. Then the patient gets into bed at 10 p.m. and says, I'm not tired. We shouldn't be surprised.

The American Academy of Sleep Medicine's stimulus-control approach includes refraining from daytime napping, while sleep restriction is specifically designed to increase sleep drive and consolidate nighttime sleep. Some CBT-I programs allow carefully timed short naps in selected circumstances—but that's very different from telling a patient with chronic insomnia to start taking 90+minute nap.

The participants in this study were acutely sleep deprived. That is not the same clinical problem.

I think this distinction gets lost sometimes. A patient who cannot sleep despite having adequate opportunity to sleep is different from a patient who simply doesn't have enough opportunity to sleep. One may have chronic insomnia. The other may be sleep deprived because life has become incompatible with adequate sleep. And increasingly, we're seeing plenty of the second group.

If someone occasionally has an unavoidable period of major sleep loss—a night-shift worker, a caregiver, a parent with a sick child—the intervention may not necessarily begin with another medication. Maybe part of the plan is behavioral: movement, exercise, morning light, protecting the next sleep opportunity, and then figuring out how to prevent the cycle from becoming chronic.

This is also where psychiatry needs to be thoughtful. Someone sleeps poorly. They're exhausted the next day. They compensate with caffeine. Then they're wired later and can't sleep. They take one of the worst over the counter medications on the planet known as diphenhydramine. Now they're groggy the following morning. More caffeine. Repeat.

For older adults, that diphenhydramine is particularly concerning because of its substantial anticholinergic burden. Suddenly a sleep problem has become a medication problem.

There are certainly times when medication is appropriate. But before adding another drug, it is worth asking what physiology we're actually trying to fix. Is this insomnia? Circadian misalignment? Insufficient sleep opportunity? Medication timing? Sleep apnea? Or simply a person whose life currently requires more waking hours than their biology can comfortably provide? Those are all very different problems.

The takeaway from this study isn't that exercise can replace sleep. It can't. And it certainly isn't that we should start recommending 90-minute naps to everyone with insomnia.

The interesting finding is that 20 minutes of exercise helped protect memory after severe sleep loss without increasing fatigue. For me, that adds another reason to think of exercise as something more specific than the generic advice we sometimes give at the end of an appointment: "Try to exercise more."

Exercise can be part of the treatment plan. For chronic insomnia, CBT-I remains the foundation, with exercise sitting where it belongs, as an adjunct rather than a replacement. But for the patient sitting across from us saying, "I barely slept, and I have to function today," we may have another tool worth discussing.

Sometimes the answer to a bad night of sleep isn't another pill. It might be a pair of sneakers.

Frimpong, E., M. Mograss, J. Dautet-Plourde, et al. "Slow-Wave Sleep Mediates the Compensatory Effects of Acute Moderate-Intensity Exercise on Declarative Memory Deficits in Sleep Restriction." Sleep, 2026.

Lotlikar, M. S., B. Ayotte, A. Choi, et al. "Protecting Episodic Memory After Sleep Loss: Similar Benefits of Exercise and Naps via Distinct Neural Contributions." Proceedings of the National Academy of Sciences of the United States of America, 2026.

Matheson, E. M., B. D. Brown, and A. O. DeCastro. "Treatment of Chronic Insomnia in Adults." American Family Physician, 2024.

Saadh, M. J., A. Y. Saleh, R. A. Kareem, et al. "Exercise as a Therapeutic Strategy for Insomnia: Current Mechanisms and Clinical Relevance." Sleep Medicine, 2025.

Wang, M., L. Su, L. Hao, et al. "Effects of Physical Exercise on Sleep Quality in Patients with Psychiatric Disorders: A Systematic Review and Meta-Analysis." BMC Psychiatry, 2026.