One study that deserves more attention than it has received.
Published earlier this year in SLEEP, the journal of the Sleep Research Society, it followed 38,838 adults and asked a deceptively simple question: what behaviours most reliably improve cardiorespiratory fitness and parasympathetic activity?
Not what supplements. Not what training protocols. What behaviours, consistently applied, produce the strongest measurable biological outcomes?
The answer was four. Morning sunlight exposure. Time-restricted eating. Zone 2 cardiovascular training. Breathwork (Holmes et al., 2026).
None of them are new.
None of them are surprising to anyone who has spent time in the performance literature.
What the study establishes, at a scale that commands serious attention, is the mechanism through which they work.
They do not directly improve cardiorespiratory fitness.
They improve it by first improving something else: sleep consistency.
That finding is worth sitting with. The behaviours themselves are not the primary lever. The consistency of the circadian signal they produce is the primary lever. Remove the consistency, and the outcomes diminish. The benefit is not in the behaviour in isolation. It is in what the behaviour does to the regularity of the biological rhythm that underlies everything else.
What the performance industry has been getting wrong
The dominant model in performance supplementation and coaching for the last two decades has been additive. Identify a deficit. Add something to address it. More protein for recovery. More caffeine for alertness. More adaptogens for stress. More nootropics for focus.
This model is not wrong precisely. Some of those additions are evidence-backed and useful. But it is incomplete in a way that compounds over time, treating the outputs of a biological system as independent variables rather than downstream consequences of how well the system is regulated.
I have spent thirty years watching this play out. When cardiorespiratory fitness declines in a high-functioning individual who trains consistently, the additive model asks what to add. The research above asks a different question: is the system producing that fitness coherent?
Is the 24-hour rhythm that governs hormonal output, cellular repair, immune function, and neural restoration running with the regularity it requires?
The distinction matters because the interventions are completely different. Adding a recovery compound to an incoherent circadian system does not render it coherent. It adds to the output side of the equation while the input side continues to produce less than it could.
Why sleep consistency outperforms sleep duration
Before getting into the four behaviours themselves, there is a prior point that the research forces us to address.
Most of the conversation about sleep and performance centres on duration. Eight hours. Seven hours. The debate about whether six is sufficient. This framing is wrong, not completely, but materially, in ways that lead to poor decisions.
Sleep regularity predicts health outcomes more strongly than sleep duration does.
Windred et al. (2024) tracked more than 60,000 UK Biobank participants using over 10 million accelerometer-hours of data. Individuals with the most irregular sleep-wake patterns had between 20 and 48 per cent higher all-cause mortality risk compared to the most regular sleepers, independent of total sleep duration. Huang, Mariani, and Redline (2020) used data from the Multi-Ethnic Study of Atherosclerosis and found that sleep-timing variability was associated with an adjusted hazard ratio of 2.11 for incident cardiovascular events. Chaput et al. (2025) confirmed this in 72,269 UK adults: meeting sleep-duration guidelines did not offset the elevated cardiovascular risk among the most irregular sleepers.
You can sleep eight hours and still be causing significant biological damage if you do so at different times each night.
That is the context in which the four behaviours become meaningful. They are not primarily sleep aids. They are circadian regulators. They work by stabilising the biological rhythm, and it is the stability of that rhythm, more than any other single variable, that determines the baseline from which all performance is produced.
The four behaviours - what they really do
Light exposure from morning sunlight
While there is a vitamin D component to this behaviour, this is largely a timing behaviour.
Photoreceptors in the retina contain specialised cells known as intrinsic photosensitive retinal ganglion cells (iprgcs). These photoreceptor cells express melanopsin. As light activates the melanopsin in the ip rgcs in the morning, they deliver a phase-stabilising signal to the suprachiasmatic nucleus (SCN); also known as the brain's central circadian pacemaker. The SCN controls everything downstream, including cortisol and melatonin production, body temperature, and hormonal cascades that regulate alertness, recovery, and metabolic function. (Berson et al., 2002; Hattar et al., 2002)
Khalsa et al. (2003) found using phase-response curve work that morning light pulses produce the largest phase advances in the human circadian clock. The effect produced by morning light pulsing is dependent upon the time of day relative to your current circadian phase.
For practical dosage recommendations, Brown et al. (2022) suggest a minimum of 250 melanopic equivalent daylight illuminance (EDI) at the eye during the day. Most indoor environments do not approach this level. However, even on overcast days, Outdoor morning light typically does. This is the single most cost-effective circadian intervention available to anyone.
The mechanism
Your eyes contain a small population of cells whose only job is to detect light and tell your brain what time of day it is. They don't help you see, they synchronise your internal clock. Morning light hits these cells and sets the entire 24-hour system: when cortisol rises, when melatonin falls, when your body temperature peaks. Miss this signal and the rest of the system runs on a guess.
Timed-restricted eating
The SCN is not the only biological clock in the body. There are molecular oscillators in every tissue and organ which run a 24-hour feedback loop of gene expression and protein synthesis. The SCN coordinates these through neural signals and hormones. However, peripheral clocks are extremely sensitive to another zeitgeber: the timing of meals.
Damiola et al. (2000) demonstrated clearly that, twenty-five years later, restricting food intake to the active phase of the day completely reverses the circadian phase shift of gene expression in the liver, kidney, heart, and pancreas, while leaving the SCN unchanged. The liver does not care what time the brain thinks it is. It responds when food arrives.
Wehrens et al. (2017) showed the human version of this: a five-hour delay in meal timing shifted the phase of peripherally measured circadian markers approximately one hour without shifting plasma melatonin, a marker of CNS activity. Both feeding timing and light timing act on different parts of the circadian system. When they are consistent with each other, the entire system runs in cohesion. When they are mismatched, peripheral clocks drift out of phase with the CNS pacemaker. The metabolic consequences can be measured clinically (Chellappa et al., 2021).
Therefore, eating within a consistent window aligned with the active phase of the day is not a weight-loss strategy, but it does offer added benefits for the circadian system. Rather, it is a circadian intervention with many potential health benefits, including weight reduction.
The mechanism
Your brain has a master clock, but every organ, liver, pancreas, gut, heart, also runs its own local clock. The brain's clock is set by light. The organ clocks are set by food. If you eat at 11pm, your liver thinks it's daytime, even though your brain knows it's night. The two systems fall out of sync, and metabolism degrades. Eating in a consistent window each day keeps the organ clocks pointing in the same direction as the brain clock.
Zone 2 Cardiovascular Training
Holloszy (1967) established the biochemical basis for aerobic training by showing that the increase in mitochondrial content in skeletal muscle due to endurance training was nearly doubled over resting levels. Higher mitochondrial density supports recovery of the parasympathetic branch of the ANS via increased HRV and enhances the autonomic flexibility, this indicates.
One important point of clarification is needed: the idea that Zone 2, specifically, rather than some form of aerobic work, has a greater effect on mitochondria is disputed in the original research. MacInnis and Gibala (2017) reported similar improvements in mitochondria in response to continuous low- to moderate-intensity training and high-intensity interval training at equal volumes. Therefore, the benefits of Zone 2 may be more related to its ability to provide regularity, recoverability, and consistency than to being a superior mechanism.
That reasoning also relates to what we are talking about here: how much can be completed and then recovered from, while avoiding sleep disruption. Zhang, Bi, and Luo (2025) conducted a meta-analysis that demonstrated that long-term aerobic-based exercise programs significantly improved HRV associated with vagal control and reduced sympathetic predominance. While the mechanism exists, the specificity of the zone 2 described is not as clearly defined as the popular story suggests.
Thomas et al. (2020) demonstrated that 5 mornings of exercise advanced the timing of the circadian rhythm, another advantage of establishing a routine when timing aligns with your body's morning activation cycle. When you follow this schedule consistently, it will reinforce the signals you send to your circadian rhythms. Again, consistency is key.
Breath Work
The mechanism of breathwork primarily affects the ANS. Breathe pacing of approximately 0.1 Hz or roughly 6 breaths/minute maximises RSA and activates the baroreceptor reflex to augment parasympathetic influence (Lehrer & Gevirtz, 2014; Bernardi et al., 2001).
Laborde et al. (2022) reviewed all studies of voluntary slow-paced breathing and found significant increases in HRV under vagal control immediately after and for multiple sessions following slow-paced breathing interventions. Fincham et al. (2023) meta-analysed 12 randomised controlled trials and found a statistically significant reduction in physiological stress markers, such as cortisol, as measured by breathwork.
Tsai et al. (2015) found that a 20-minute presleep slow-paced breathing intervention decreased sleep onset time and improved overall sleep quality among individuals with chronic insomnia.
Again, I need to clarify exactly what type of effect that breathwork is having here versus what it isn't. Breathwork doesn't shift the circadian rhythm phase in the same way as light exposure.
There is limited research on whether breathwork directly affects sleep duration as an outcome measure (sleep quality within a single night is a different concept); however, there is evidence that breathwork modulates the autonomic state in a way that enhances the parasympathetic shift necessary for good sleep onset and recovery. In fact, this is very closely related to but distinct from circadian modulation.
The mechanism
Breathing at roughly six breaths per minute, five seconds in, five seconds out, activates a reflex loop between your heart and the pressure sensors in your arteries. That loop is the body's main mechanism for shifting from "alert" to "recover" mode. Twenty minutes of this before sleep does something a hot bath or a meditation app cannot: it directly trains the nervous system to make the transition that good sleep requires.
The supplement industry's missing variable
Most supplement systems are designed around what happens inside a single serving. The compound is selected, the dose is calibrated, and the delivery mechanism is chosen. What is rarely designed is the relationship between the supplement, the time of day at which it is taken, and the biological state present at that moment.
This is not a marginal consideration.
Chronopharmacology, the study of how the timing of drug administration affects efficacy and safety, demonstrates that the same compound can produce materially different effects at different circadian phases, because the receptor systems it interacts with, the enzymatic pathways it enters, and the hormonal environment it encounters all vary across the 24-hour cycle (Dallmann, Brown and Gachon, 2014; Dallmann, Okyar and Lévi, 2016).
This is not theoretical. Awad et al. (2017) meta-analysed eleven RCTs and found that evening dosing of short-half-life statins reduced LDL cholesterol by an additional 9.68 mg/dL compared to morning dosing, a clinically meaningful difference produced solely by the timing of the same compound. Long et al. (2016) showed that morning influenza vaccination produced significantly higher antibody titres than afternoon vaccination in adults over 65.
These are pharmaceutical examples, not supplement examples. The direct RCT evidence for chronotype-adjusted supplementation with nutrients and performance compounds is limited. But the mechanism is the same. Pharmacokinetics does not operate in a biological vacuum. It operates within a system whose state changes over time of day.
A supplement designed without reference to circadian phase is a supplement designed for an average biological state that no individual is ever actually in.
The mechanism
The same compound, given at a different time of day, can produce a noticeably different effect. Statins lower cholesterol more in the evening. Flu vaccines produce a stronger immune response in the morning. The drug hasn't changed, the biological state it lands in has. This is why a single multivitamin taken whenever is a different intervention to compounds matched to the biological phase they're entering.
What the zeitgeber hierarchy tells us
One last layer warrants our attention before we move on to how these behaviours can be applied.
The four behaviours described by Holmes et al. Were not developed as separate "equal" levers. Instead, each behaviour sits atop a hierarchy of zeitgebers, environmental and behavioural stimuli that entrain biological clocks. This hierarchy is organised in two different ways depending upon whether you're looking at the central clock or your peripheral clocks.
For example, light is the dominant zeitgeber for the SCN. It is the primary timekeeper that synchronises the central circadian pacemaker. No other input has an opportunity to compete with it in healthy-eyed humans.
For peripheral clocks (the metabolic tissues), feeding timing is the dominant zeitgeber. These metabolic tissues are most responsive to when (not what) you eat. Exercise and social cues serve as weak secondary zeitgebers. When timed consistently with the central signal, they will support its influence. However, when timed inconsistently with the central signal, they may create conflict (Roenneberg, Daan & Merrow, 2003).
When zeitgebers conflict, when you eat at midnight, train at random times, and get inadequate Morning light, your peripheral clocks will drift out of phase with your central pacemaker. A study by Heyde & Oster (2019) demonstrates that conflicting zeitgeber conditions result in measurable declines in metabolism compared with synchronous conditions. The cost is physiologically real.
The four behaviours work because they are inputs to this same hierarchy. For example, Morning light reinforces the central clock. Timely meals reinforce your peripheral clocks. Regular Morning exercise adds another reinforcing signal to the central system. Breathwork supports your evening parasympathetic shift, allowing the system to recover cleanly. Each behaviour has a specific position in the zeitgeber architecture. That is not a coincidence. That is the mechanism
The actions, applied
The research points to four behaviours. Here is what applying them actually looks like.
Get outside within the first hour of waking.
Not a lamp (unless it provides biologically significant light). Not a screen. Outdoor light, even on an overcast day, delivers melanopic stimulus that indoor environments almost never match. The target is 250 melanopic EDI at the eye, the consensus threshold from Brown et al. (2022). Ten to thirty minutes is sufficient on a clear day. The earlier relative to your natural wake time, the stronger the phase-stabilising signal.
Eat within a consistent window, aligned with the day.
The research does not prescribe a specific window length. It prescribes consistency. The liver, pancreas, and gut contain molecular clocks that respond to the arrival of food. Irregular eating timing drifts those peripheral clocks out of phase with your central one. A consistent eight- to ten-hour eating window, starting within two hours of waking, provides peripheral clocks with a secondary zeitgeber that reinforces rather than contradicts the morning light signal.
Train aerobically at a low-to-moderate intensity regularly, ideally in the morning.
The mitochondrial and autonomic benefits of aerobic training are well established and not intensity-dependent, as the Zone 2 narrative sometimes implies. What matters is consistency and recoverability. Training you can do four to five times a week without disrupting sleep is more valuable than training intensity that compromises it. Morning timing adds a phase-advancing circadian benefit on top of the autonomic one.
Use breathwork in the evening, not as a relaxation technique but as an autonomic intervention.
Paced breathing at approximately six breaths per minute, five seconds in, five seconds out, activates the baroreflex in a way that directly augments parasympathetic tone. Twenty minutes before sleep. This is not meditation. It is a specific, dosed input to the autonomic nervous system at the phase of the day when the biology is attempting to make the sympathetic-to-parasympathetic transition that clean sleep onset requires.
The common thread across all four is this: each behaviour is a timed, consistent input to a system that responds to timing and consistency above all else. None of them works at full effect when applied sporadically. All of them compound applied with regularity. That is not motivational language. That is the mechanism the research identifies.
What this means in practice
The Holmes et al. (2026) paper is notable not only for its findings but for what it establishes about the mechanism. In 38,838 adults, across four distinct behavioural inputs, the pathway to improved cardiorespiratory fitness and parasympathetic activity ran through a single common variable: sleep consistency. That is not a minor footnote. That is the finding.
It is worth noting that this is a large-scale observational analysis rather than a randomised controlled trial, and the authors are transparent about that. But the mechanistic argument it makes is independently corroborated by prospective cohort data carrying hard endpoints. Windred et al. (2024) found a 20 to 48 per cent reduction in all-cause mortality risk among the most regular sleepers in a cohort of over 60,000 adults. Huang et al. (2020) reported a hazard ratio of 2.11 for incident cardiovascular events in the most sleep-irregular group. Chaput et al. (2025) confirmed that meeting sleep-duration guidelines did not offset elevated cardiovascular risk among irregular sleepers in 72,269 UK adults.
The Holmes paper sits within a body of evidence that is accumulating rapidly and pointing consistently in the same direction. Sleep consistency is not a proxy for good habits in general. It is a primary input to the biological system that produces the capacity most high-functioning people are trying to protect and extend.
The practical implication follows directly from that. For the individual running on inconsistent schedules, long-haul travel, and a stimulant load calibrated to compensate for an underlying timing problem, the four behaviours are a map back to the system. They are not a protocol to add on top of existing habits. They are the structural foundation on which everything else, including any well-designed supplementation, builds.
The industry is beginning to understand this. The question now is whether the products and systems available to high-functioning people are built with that understanding, or built for a model the evidence is leaving behind.
At HMN24, the architecture of RISE, FLOW, and PRE-SLEEP was built around the same mechanism this research identifies: timed, consistent inputs to a biological system that runs on timing and consistency. Not more stimulation. Correctly timed biological support, across the full 24 hours.
The behaviours come first. The system is built to work with them.
References
Awad, K., Serban, M-C., Penson, P., Mikhailidis, D.P., Toth, P.P., Jones, S.R., Rizzo, M., Howard, G., Lip, G.Y.H. and Banach, M. (2017) 'Effects of morning vs evening statin administration on lipid profile: a systematic review and meta-analysis', Journal of Clinical Lipidology, 11(4), pp. 972–985.
Bernardi, L., Sleight, P., Bandinelli, G., Cencetti, S., Fattorini, L., Wdowczyc-Szulc, J. and Lagi, A. (2001) 'Effect of rosary prayer and yoga mantras on autonomic cardiovascular rhythms: comparative study', BMJ, 323(7327), pp. 1446–1449.
Berson, D.M., Dunn, F.A. and Takao, M. (2002) 'Phototransduction by retinal ganglion cells that set the circadian clock', Science, 295(5557), pp. 1070–1073.
Brown, T.M., Brainard, G.C., Cajochen, C., Czeisler, C.A., Hanifin, J.P., Lockley, S.W., Lucas, R.J., Münch, M., O'Hagan, J.B. and Peirson, S.N. et al. (2022) 'Recommendations for daytime, evening, and nighttime indoor light exposure to best support physiology, sleep, and wakefulness in healthy adults', PLOS Biology, 20(3), e3001571.
Chaput, J-P., Biswas, R.K., Ahmadi, M., Cistulli, P.A., Rajaratnam, S.M.W., Bian, W., St-Onge, M-P. and Stamatakis, E. (2025) 'Sleep regularity and major adverse cardiovascular events: a device-based prospective study in 72,269 UK adults', Journal of Epidemiology and Community Health, 79(4), pp. 257–264.
Chellappa, S.L., Qian, J., Vujovic, N., Morris, C.J., Nedeltcheva, A., Nguyen, H., Rahman, N., Heng, S.W., Kelly, L., Kerlin-Monteiro, K. et al. (2021) 'Daytime eating prevents internal circadian misalignment and glucose intolerance in night work', Science Advances, 7(49), eabg9910.
Dallmann, R., Brown, S.A. and Gachon, F. (2014) 'Chronopharmacology: new insights and therapeutic implications', Annual Review of Pharmacology and Toxicology, 54, pp. 339–361.
Dallmann, R., Okyar, A. and Lévi, F. (2016) 'Dosing-time makes the poison: circadian regulation and pharmacotherapy', Trends in Molecular Medicine, 22(5), pp. 430–445.
Damiola, F., Le Minh, N., Preitner, N., Kornmann, B., Fleury-Olela, F. and Schibler, U. (2000) 'Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus', Genes and Development, 14(23), pp. 2950–2961.
Dibner, C., Schibler, U. and Albrecht, U. (2010) 'The mammalian circadian timing system: organisation and coordination of central and peripheral clocks', Annual Review of Physiology, 72, pp. 517–549.
Fincham, G.W., Strauss, C., Montero-Marin, J. and Cavanagh, K. (2023) 'Effect of breathwork on stress and mental health: a meta-analysis of randomised-controlled trials', Scientific Reports, 13(1), 432.
Hattar, S., Liao, H.W., Takao, M., Berson, D.M. and Yau, K.W. (2002) 'Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity', Science, 295(5557), pp. 1065–1070.
Heyde, I. and Oster, H. (2019) 'Differentiating external zeitgeber impact on peripheral circadian clock resetting', Scientific Reports, 9(1), 20114.
Holmes, K.E., Kim, J., Fielding, F., Zeitzer, J.M. and von Hippel, W. (2026) 'Four core circadian behaviours that improve cardiorespiratory fitness through consistent sleep', SLEEP, 49(2), zsaf318.
Holloszy, J.O. (1967) 'Biochemical adaptations in muscle: effects of exercise on mitochondrial oxygen uptake and respiratory enzyme activity in skeletal muscle', Journal of Biological Chemistry, 242(9), pp. 2278–2282.
Huang, T., Mariani, S. and Redline, S. (2020) 'Sleep irregularity and risk of cardiovascular events: the Multi-Ethnic Study of Atherosclerosis', Journal of the American College of Cardiology, 75(9), pp. 991–999.
Khalsa, S.B.S., Jewett, M.E., Cajochen, C. and Czeisler, C.A. (2003) 'A phase response curve to single bright light pulses in human subjects', The Journal of Physiology, 549(3), pp. 945–952.
Laborde, S., Allen, M.S., Borges, U., Dosseville, F., Hosang, T.J., Iskra, M., Mosley, E., Salvotti, C., Spolverato, L., Zammit, N. and Javelle, F. (2022) 'Effects of voluntary slow breathing on heart rate and heart rate variability: a systematic review and a meta-analysis', Neuroscience and Biobehavioural Reviews, 138, 104711.
Lehrer, P.M. and Gevirtz, R. (2014) 'Heart rate variability biofeedback: how and why does it work?', Frontiers in Psychology, 5, 756.
Long, J.E., Drayson, M.T., Taylor, A.E., Toellner, K.M., Lord, J.M. and Phillips, A.C. (2016) 'Morning vaccination enhances antibody response over afternoon vaccination: a cluster-randomised trial', Vaccine, 34(24), pp. 2679–2685.
MacInnis, M.J. and Gibala, M.J. (2017) 'Physiological adaptations to interval training and the role of exercise intensity', The Journal of Physiology, 595(9), pp. 2915–2930.
Phillips, A.J.K., Clerx, W.M., O'Brien, C.S., Sano, A., Barger, L.K., Picard, R.W., Lockley, S.W., Klerman, E.B. and Czeisler, C.A. (2017) 'Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing', Scientific Reports, 7(1), 3216.
Roenneberg, T., Daan, S. and Merrow, M. (2003) 'The art of entrainment', Journal of Biological Rhythms, 18(3), pp. 183–194.
Thomas, J.M., Kern, P.A., Bush, H.M., McQuerry, K.J., Black, W.S., Clasey, J.L. and Pendergast, J.S. (2020) 'Circadian rhythm phase shifts caused by timed exercise vary with chronotype', JCI Insight, 5(3), e134270.
Tsai, H.J., Kuo, T.B.J., Lee, G.S. and Yang, C.C.H. (2015) 'Efficacy of paced breathing for insomnia: enhances vagal activity and improves sleep quality', Psychophysiology, 52(3), pp. 388–396.
Wehrens, S.M.T., Christou, S., Isherwood, C., Middleton, B., Gibbs, M.A., Archer, S.N., Skene, D.J. and Johnston, J.D. (2017) 'Meal timing regulates the human circadian system', Current Biology, 27(12), pp. 1768–1775.
Windred, D.P., Burns, A.C., Lane, J.M., Saxena, R., Rutter, M.K., Cain, S.W. and Phillips, A.J.K. (2024) 'Sleep regularity is a stronger predictor of mortality risk than sleep duration: a prospective cohort study', Sleep, 47(1), zsad253.
Zhang, W., Bi, S. and Luo, L. (2025) 'The impact of long-term exercise intervention on heart rate variability indices: a systematic meta-analysis', Frontiers in Cardiovascular Medicine, 12, 1364905.
