The temperature lever

For most of human history, deliberate exposure to extreme heat or cold was a practical matter. You had a sauna because Finland is cold and bathing in winter required heat. You jumped in a lake because the lake was there. Nobody was tracking biomarkers.

In the last decade, both practices moved to the center of a popular wellness culture, mostly through two figures: Wim Hof and Andrew Huberman. Cold plunges went from a Scandinavian eccentricity to a Silicon Valley protocol. Saunas, somewhat quieter in the noise, picked up serious longevity-research backing. Both are now claimed to do almost anything: boost testosterone, reverse depression, prevent dementia, increase metabolism, train resilience, kill cancer cells, optimize hormones, sharpen focus.

Most of these claims are either oversold or wrong. A few are well-evidenced. The honest reader needs a way to tell which is which.

Here is the short version, which the rest of this piece unpacks:

• Sauna has the stronger evidence by a wide margin. The Finnish KIHD cohort, followed for over twenty years, shows large reductions in cardiovascular mortality, sudden cardiac death, and dementia. The biological mechanisms are mapped. Sauna is closer to a "discovered intervention with real effects" than to wellness hype.
• Cold exposure has real but smaller effects. The acute neurochemistry (large norepinephrine and dopamine surges) is well-documented. The downstream claims (immunity, weight loss, depression, testosterone) are largely overstated or weakly evidenced. The "feels great after" effect is real. Most of the rest is influencer overhang.
• They are different practices for different effects. Treating them as a single category ("temperature exposure") obscures what each one is doing.
• Both have real safety issues that the marketing systematically downplays.

This piece is shorter than the others in the series because it should be. Temperature is a narrower topic with less surface area. The goal is to give you enough to choose well, not to pad.

Sauna is medicine with a long tail. Cold is a tool with a sharp edge. Different tools, different jobs.

Section 1Heat: the Finnish evidence

The cardiovascular case for sauna is the strongest single body of evidence on any practice in this series, sleep included.

The Kuopio Ischaemic Heart Disease Risk Factor Study (KIHD) followed 2,315 middle-aged Finnish men for a median of twenty-one years. Jari Laukkanen and colleagues published the headline result in JAMA Internal Medicine in 2015. Men who used a traditional Finnish sauna 4 to 7 times per week, compared to men using it once per week, had a 40% lower all-cause mortality, a 63% lower risk of sudden cardiac death, and a 50% lower risk of fatal cardiovascular disease. The relationship was dose-dependent: 2 to 3 sessions per week produced about a 23% reduction; 4 to 7 sessions produced the larger effect. Longer sessions (over 19 minutes) outperformed shorter ones.

A 2017 follow-up in Age and Ageing extended the finding to dementia: the 4-to-7-times-per-week group had a 66% lower risk of any dementia and a 65% lower risk of Alzheimer's disease. Subsequent papers from the same cohort have shown associations with reduced hypertension, lower stroke risk, and improved blood pressure trajectories.

The evidence is observational, not randomized, so the usual caveats apply: people who can use saunas frequently might be healthier to begin with. Healthy-user bias is real. But the effect sizes are unusually large, the dose-response is clean, the mechanisms are biologically plausible, and the same cohort has shown the pattern across multiple cardiovascular and neurodegenerative endpoints. This is about as strong as observational data gets without a randomized trial.

What's actually happening in the body

Sitting in 80 to 100 degrees Celsius (roughly 175 to 210 Fahrenheit) for 15 to 30 minutes is a controlled physiological stress. Several things happen at once:

• Cardiovascular load similar to moderate exercise. Heart rate climbs to 120 to 150 beats per minute. Cardiac output rises by 60 to 70 percent. Peripheral vasodilation drops systolic blood pressure, then the body adapts over weeks of regular use by lowering resting blood pressure modestly. The cardiovascular system trains the same way it does on a bike, without the joint impact.
• Heat shock proteins. When core temperature rises by about 1 degree Celsius, cells upregulate a family of chaperone proteins (HSP70, HSP90 and others) that stabilize and refold damaged proteins. A single 30-minute session at 73 degrees Celsius increases HSP levels roughly 50% above baseline. Heat shock proteins are involved in cellular stress resistance, protein quality control, and longevity pathways. The HSP90/eNOS/nitric oxide axis is the most-studied molecular pathway linking sauna to endothelial health.
• Endothelial conditioning. Repeated heat exposure improves the function of the vascular endothelium, the single-cell lining of blood vessels that regulates tone, clotting, and inflammation. Endothelial dysfunction is upstream of most cardiovascular disease. This is probably one of the major mechanisms behind the mortality data.
• Mild glymphatic stimulation and elevated growth hormone after sessions. Modest but real.
• Subjective mood effect. Acute parasympathetic rebound and beta-endorphin release. People feel good after a sauna for the same reason they feel good after a sustained workout.

Dose, temperature, and what counts

The KIHD data is for traditional dry Finnish sauna, with a hot rock heater, ambient temperature 80 to 100 Celsius, sessions of 15 to 30 minutes, optionally with some steam (löyly) from water poured on the rocks. This is the form on which the evidence rests.

Infrared sauna is not the same thing. Infrared cabins typically heat to 45 to 65 Celsius and warm the body directly through radiant energy rather than through ambient air temperature. Core temperature rises more slowly and less reliably. The cardiovascular load is lower. There is some research on infrared sauna for chronic pain and rheumatoid arthritis, but the long-term cardiovascular and dementia data does not exist for infrared at anything close to the depth that exists for traditional Finnish sauna. Treating them as interchangeable is a category error. If your option is an infrared cabin, use it (it's not nothing) but do not assume the Finnish mortality data transfers cleanly.

Steam rooms and Turkish hammams are also distinct: cooler ambient temperature (40 to 50 Celsius) with high humidity. They produce real cardiovascular load but again, the long-term cohort evidence is for dry Finnish sauna specifically.

The honest summary on sauna

If you can use a traditional Finnish sauna 3 or 4 times per week for 15 to 30 minutes, the evidence says you are probably reducing cardiovascular and dementia risk meaningfully, with effects in the same magnitude as regular aerobic exercise. The mechanisms are real, the data is unusually clean for an observational study, and the practice has been tested at population scale for generations. It is, on a cost-benefit basis, one of the higher-leverage long-term-health interventions available, provided you have access to one.

Section 2Cold: what's actually there

The cold side of the story is more interesting and more crowded with bad claims. Let's start with what's real.

Acute neurochemistry

The Šrámek et al. 2000 paper (European Journal of Applied Physiology) is the foundational reference. They immersed healthy volunteers in 14 degree Celsius water for an hour and measured catecholamines. Norepinephrine rose by roughly 530 percent. Dopamine rose by about 250 percent. These are very large changes, sustained over the hour. Cortisol changed less.

Multiple follow-up studies have confirmed the pattern at shorter durations and slightly different temperatures. Even a 3-to-5-minute cold immersion at 10 to 14 Celsius produces a measurable, sustained catecholamine surge. This is the underlying mechanism behind the "feels alert for hours after" effect. Norepinephrine in particular has a long half-life in the periphery and stays elevated well past the immersion itself.

Mood and subjective state

The acute mood improvement after cold exposure is the most consistently reported subjective effect, both in the lab and in self-report. The 2025 systematic review and meta-analysis of cold-water immersion (11 RCTs, 3,177 participants) found measurable but modest effects on mood and stress reactivity. The dopamine elevation almost certainly contributes; so does the parasympathetic rebound that follows the initial sympathetic burst, similar to the rebound after intense exercise. The "I feel weirdly good for two hours after" experience is real and pharmacologically explicable.

Whether this translates into durable improvements in mood disorders is the harder question. A handful of small studies and case reports suggest cold-water swimming may help in treatment-resistant depression, but the evidence is preliminary, the studies are tiny, and the effect is conflated with the social and physical-activity aspects of, say, joining a swimming club. The honest reading is: maybe useful as an adjunct, far too early to call it a treatment.

Brown fat activation

Cold exposure activates brown adipose tissue (BAT), a metabolically distinct fat tissue that burns calories as heat rather than storing them. This is real. The Cypess and van Marken Lichtenbelt labs have shown that two hours at 19 Celsius measurably raises energy expenditure and BAT activity in humans, with cold acclimation over weeks recruiting more BAT.

The honest version is that the metabolic effect on humans is small. Cold-induced thermogenesis at moderate exposures adds on the order of 100 to 250 kcal per day in the best studies, and most of that requires sustained mild cold (hours, not minutes). The "ice baths burn fat" framing in popular content overstates this by an order of magnitude. BAT activation is biologically interesting; it is not a practical weight-loss tool.

The vagal lever (cold on the face)

Brief cold to the face triggers the mammalian dive reflex: cold receptors around the forehead, eyes, and cheeks activate a vagal response that slows the heart within seconds. This is the fastest non-drug way to drop acute sympathetic activation. Splashing cold water on your face when anxious works for this specific reason, and you don't need full immersion to get the effect. The upward-spiral piece covers this lever in more detail.

The Wim Hof endotoxin study

The most-cited "cold boosts immunity" finding is Kox, Pickkers and colleagues at Radboud (2014, PNAS). Twelve healthy men trained for ten days in the full Wim Hof Method (cyclic hyperventilation, breath holds, daily cold exposure). They were then injected with bacterial endotoxin. The trained group showed lower inflammatory markers (TNF-alpha, IL-6, IL-8), higher anti-inflammatory IL-10, fewer flu-like symptoms.

This is a real finding. But two caveats matter. First, it is small (24 men total, 12 per group), with no replication of equivalent quality, and the protocol bundles breathing, cold, and meditation, so it is not clean evidence for cold specifically. Most of the immune-suppression effect appears to come from the breath protocol (sympathetic activation via hyperventilation, adrenaline release), not the cold. The breath piece goes deeper into this.

The honest version: there is a signal that voluntary sympathetic activation can attenuate acute inflammatory response, the WHM protocol does this, but generalizing from "WHM attenuated one experimental endotoxin challenge" to "cold plunges boost your immune system" is a long leap that the data does not support.

The resilience argument

The hardest-to-measure but most defensible claim about cold exposure is the stress-inoculation framing. Voluntarily exposing yourself to a controllable acute stressor, surviving it, and rebounding may train the nervous system to handle other stressors better. This is the argument the stress-response piece calls "training the recovery." There is reasonable theoretical support for this from animal models of hormesis and human work on stress reactivity, but direct evidence specifically for cold plunges as resilience training is thin. Most people who report this effect are conflating the catecholamine surge with the long-term adaptation. The two might overlap. They are not the same thing.

Section 3Cold: where the claims overreach

The cold-plunge culture has accumulated several claims that the evidence does not support, or contradicts.

Testosterone

"Cold plunges boost testosterone" is one of the most common claims and it is not supported. No peer-reviewed study that directly measures testosterone has found a meaningful, lasting increase from cold water immersion. The 2025 study of habitual winter swimmers found a decrease in testosterone after immersion. Acute scrotal cooling does briefly raise testosterone, but the magnitude is small and transient. As a strategy for raising baseline testosterone, this is not real. The original "cold boosts testosterone" claims in influencer culture appear to have generalized from animal data, scrotal-cooling fertility studies, or wishful extrapolation. Save your nuts.

Weight loss

The BAT calorie math is what it is. A few hundred calories of cold-induced thermogenesis per day, only with sustained exposure, is not a useful weight-loss tool. People lose weight after starting cold protocols mostly because they also change their diet, exercise more, and pay more attention to their body. The cold itself contributes very little.

Depression

Some preliminary case reports and very small studies. No large RCTs. Worth trying as a low-cost adjunct for the right person. Not a treatment. Anyone with severe or treatment-resistant depression should not be doing this in place of evidence-based care.

Hypertrophy: the post-lift cold timing problem

This is the one cold-related finding that is well-replicated and matters for a specific group of people: anyone doing strength training.

Roberts et al. (2015, Journal of Physiology) had 21 active men strength-train twice per week for 12 weeks. Half did 10 minutes of cold water immersion (10 Celsius) after each session, half did active recovery. The cold-immersion group ended up with measurably worse adaptations:

• Type II muscle fiber cross-sectional area: 17% larger in the active-recovery group.
• Myonuclei per fiber: 26% higher in active recovery.
• Acute mTOR signalling and satellite cell activity (the molecular machinery of muscle growth): blunted by cold immersion.

A 2019 Fyfe et al. replication in the Journal of Applied Physiology found similar attenuation of hypertrophy, even though pure strength gains were preserved. Multiple subsequent reviews confirm the pattern.

The mechanism is straightforward. Strength training works by creating an acute inflammatory and anabolic signal that the body responds to over hours. Cold immersion right after lifting damps that signal. You measurably grow less muscle.

The Wim Hof bundle problem

A lot of "cold exposure benefits" research is really Wim Hof Method research, which bundles breathing, cold, and meditation. The breathing component is doing most of the heavy lifting on the immune findings. The cold component is doing the catecholamine surge and the subjective alertness. Treating the bundle as evidence for cold alone is bad reasoning. The breath piece covers this in detail.

Section 4Heat and cold are not interchangeable

One of the lazier framings in popular content is "temperature exposure" as a single category. The mechanisms point in very different directions.

One produces dilation and a parasympathetic finish; the other produces constriction and a sympathetic spike. One is a cardiovascular workout; the other is a neurochemical shock. The downstream effects barely overlap.

The implication: choose the practice for the effect you want. Long-term cardiovascular and cognitive risk reduction, with a relaxing finish? Sauna. A morning energy hit, alertness, possibly a resilience-training rep? Cold. Treating them as interchangeable is like treating squats and running as interchangeable because both involve legs.

Section 5The contrast question

The Scandinavian sauna-and-plunge sequence is centuries old and feels excellent. The mechanism is clear in the short term: peripheral vasodilation alternating with constriction is a vascular pumping action, and the contrast between systems leaves most people feeling unusually relaxed and clear-headed.

The long-term evidence specific to contrast (as opposed to either practice on its own) is thin. There is some signal for improved circulation, lymphatic flow, and post-exercise recovery in athletes, but the studies are small and heterogeneous. The headline cardiovascular and dementia data is for sauna alone, not the contrast cycle.

The honest position: hot-then-cold is enjoyable, safe at moderate intensities, probably good for circulation, and not separately evidenced as a high-leverage intervention. Do it if you enjoy it; do not assume contrast adds materially to what sauna alone is already doing.

Section 6Practical protocols

If you have access to a traditional sauna, that is the higher-priority practice. If you do not, the cold side is essentially free and works as an entry point. You do not need both.

Section 7Safety

Both practices have killed people. The risks are uncommon but real, and they are uneven: most are clustered around specific failure modes that are easy to name and avoid.

Sauna

Cold

The general principle

Both heat and cold are doses of physiological stress. The goal is enough stress to trigger adaptation, not enough to cause damage. Most healthy adults can handle moderate doses without issue. The line between adaptive stress and damage is real but rarely the issue if you respect duration limits, avoid alcohol, never combine cold immersion with hyperventilation, and stop when symptoms feel wrong rather than pushing through.

Extreme protocols (multi-hour ice baths, sauna marathons, sauna-then-icewater repeated to exhaustion) are where most of the real harm has occurred. There is no evidence that extreme doses are more beneficial than moderate ones. Almost all of the cohort evidence is for moderate, sustainable practice.

Section 8The honest middle

Here is what is worth remembering.

Sauna is one of the better-evidenced longevity interventions available. The Finnish cohort data is unusually clean, the mechanisms are mapped, and the practice is enjoyable and sustainable. If you have access, use it. Three to four sessions per week is enough; more is fine. Avoid alcohol, respect the contraindications, and treat it as a long-term practice rather than an occasional luxury.

Cold exposure does something real, but smaller and narrower than the influencer culture claims. The acute neurochemistry is genuine. The mood lift is genuine. The resilience-training argument is plausible. The testosterone, weight-loss, and immunity claims are mostly overhang. The hypertrophy-blunting effect after strength training is well-replicated and worth respecting. A cold shower is enough for most of the benefit. A daily plunge is fine if you enjoy it. A daily plunge that ends with strength training will measurably cost you muscle.

They are not the same practice. Treating them as two flavors of "temperature therapy" obscures what each one is doing. Heat is a cardiovascular and cellular conditioning tool with a long-term mortality signal. Cold is a neurochemical stimulant with a short-term alertness signal.

Neither is required for a healthy life. The interventions in the earlier pieces of this series (sleep, breath, light, movement, social regulation) all sit higher in the priority order. But if the foundations are in place and you want to layer something on top, sauna in particular is one of the better choices available.

Heat to live longer.
Cold to wake up.
Neither to perform a personality.