Sleep and HRV are tied together in two ways. HRV changes with each sleep stage across the night, and how much and how regularly you sleep shows up in the numbers. Both are measured. How sleep ranks against every other HRV input is not: we haven’t found a study that compares them head to head, so we don’t claim it’s the largest.
What happens to HRV during sleep
HRV is not constant across the night. It moves with sleep stage.
- NREM sleep: parasympathetic dominance. Trinder and colleagues measured 14 young healthy subjects over three nights. During non-REM sleep, autonomic balance moved toward the parasympathetic side, mostly through a rise in parasympathetic activity.
- REM sleep: closer to waking.The same study found autonomic balance during REM “in general similar to wakefulness.” A review by Tobaldini et al. (2013) describes REM as a likely sympathetic predominance with vagal withdrawal, and the opposite in non-REM sleep.
“During non-rapid eye movement (NREM) sleep autonomic balance shifted from sympathetic to parasympathetic dominance, although this appeared to be more because of a shift in parasympathetic nervous system (PNS) activity.”
Trinder’s group also found that the HF and LF components changed abruptly at sleep onset and then held constant within each stage, so shifts in autonomic balance across the night tracked the changing mix of stages. One implication, ours rather than theirs: an overnight HRV average depends partly on the mix of stages in the night, not only on how long you slept.
Sleep debt and HRV
Two patterns get conflated: one short night, and many. The evidence on both is mixed.
Acute sleep loss. A 2025 meta-analysis of 11 randomized trials found that sleep deprivation significantly lowered RMSSD and raised LF power and the LF/HF ratio; the changes in SDNN and HF power were not significant. Against that, Quintana et al. (2017) kept 20 healthy young men awake for a night and found no change in supine high-frequency HRV compared with 20 who slept normally.
“This meta-analysis indicates that sleep deprivation may impair cardiac autonomic function, as evidenced by decreased RMSSD and increased LF and LF/HF, suggesting sympathetic predominance and vagal suppression.”
Repeated restriction. Also split. Yang et al. (2018) held 21 people to about half their normal sleep in repeated blocks over a 22-day hospital stay. High-frequency HRV rebounded above baseline on the second recovery night, and the authors concluded that autonomic recovery takes at least three nights. Tobaldini et al. (2017) cut 8 people to two-thirds of their usual sleep for eight nights and found the normal pattern of autonomic control during NREM sleep preserved.
So we don’t give a percentage for what one short night, or a short week, does to your number. If your seven-day average has been low for a month and training, stress and alcohol don’t explain it, sleep is a reasonable next thing to examine.
Timing matters, not just duration
The most direct test we’ve found is small. Gonzales and Narvaez had 16 adults spend 12 nights on a regular sleep schedule and 12 on their habitual one, in random order. Average sleep duration was similar, but night-to-night variation in duration fell from 57 to 31 minutes. After the regular block, resting RMSSD was 42 ms against 34 ms, and resting heart rate 62 bpm against 65.
A field study of 33 young men by Sűdy et al. (2019) points the same way. In those with more social jetlag, the gap between workday and free-day sleep timing, a vagal HRV index was lower in the first three hours of sleep on workdays than on free days.
Practical: a consistent bedtime is the change these studies tested. How consistent is consistent enough, 16 people can’t tell us.
Alcohol and sleep
Alcohol before bed lowers HRV during sleep, and more alcohol lowers it further: in the lab (Sagawa et al., 2011) and across 4,098 people in everyday life (Pietilä et al., 2018). It isn’t a simple loss of deep sleep. A review by Ebrahim et al. (2013) found that in most studies alcohol increases slow-wave sleep in the first half of the night, while it delays REM and disrupts the second half. See the full alcohol page for the numbers.
Where breathing fits in
The evidence is small but direct. In Tsai et al. (2014), 14 people with self-reported insomnia practised slow paced breathing for 20 minutes before bed; on polysomnography, sleep onset latency, awakenings and time awake fell, and sleep efficiency rose. In a randomized trial of 64 healthy young adults, Laborde and colleagues compared 15 minutes of slow-paced breathing each evening for 30 days against the same time on social media. The breathing group reported better sleep quality and showed higher overnight high-frequency HRV.
Two limits. Neither study tested 4-7-8; both used slow paced breathing, for 15 to 20 minutes. And neither compared the effect with sleep duration itself, so we don’t rank them.
Put together, the studies on this page point toward four habits: enough sleep, a regular schedule, no alcohol close to bed, and slow breathing before sleep. They don’t tell you how the four compare, or how much each will move your number. Your own trend will.