The Verification Venue · a thing everyone half-gets-wrong
The Cold You Can't Catch From the Cold
Wet hair on a winter morning cannot give you a cold. A cold is a virus, mostly rhinovirus, and no amount of chill will conjure one out of air that has none in it. But the old wives were pointing at something real, just in the wrong order: a cold nose is a better place for a virus you have already breathed in.
The verdict
Turn the virus on and cool the nose to see it. With no virus present, nothing here produces a cold at any temperature. That is the whole point.
In 1946 the Medical Research Council opened the Common Cold Unit at Salisbury and spent decades trying to give volunteers colds with cold itself: long walks on freezing days, unheated flats, sitting around in wet socks. It never worked. Its director, Sir Christopher Andrewes, admitted: "We couldn't satisfy ourselves that chilling by itself did anything." Chilling alone cannot give you a cold. That is settled, and it is layer one.
Layer two is the twist the fifty thin explainers skip. In 2015 a Yale team (Foxman, Iwasaki et al.) grew the common-cold virus in airway cells at two temperatures: the ~33 °C of a nose breathing cool air, and the 37 °C of your core, and watched what the virus and the cell each did. Below is that experiment as an instrument. The virus must be present first.
Rhinovirus replication (vs 37 °C)
100×
a cooled nose lets it replicate far more
Interferon / ISG defense (index)
1
antiviral response is weakest when cold
Breathe cold, dry air and your nasal lining drops toward 33 °C. That is the only thing chill changes, and it changes the host, never the question of whether a virus is there.
Slide the nose warm to 37 °C and the virus is throttled roughly 100-fold while the cell's interferon alarm rings loudest. Slide it cold to 33 °C and the virus replicates most while the alarm is quietest: a friendlier host. Now flip the virus off: both curves flatten to nothing, at every temperature. No virus, no cold, cold nose or not. That flat line is the answer the myth gets backwards.
Read it straight off the numbers
The one hard measurement is the endpoint ratio: in wild-type mouse airway cells, the final viral titer was about 100× higher at 33 °C than at 37 °C (Foxman 2015, Fig. 4). The table interpolates log-linearly between those two measured endpoints; the intermediate rows are a model of the trend, not extra data points.
| nasal temp | rhinovirus replication | interferon / ISG defense | host for a virus you already have |
|---|
The check: every number recomputed in front of you
The replication curve is anchored on a single published fact and one interpolation rule. For the nose temperature you set, here is the arithmetic:
What is measured vs modelled. The 100× endpoint ratio is measured (mouse airway cells, mouse-adapted rhinovirus 1B). Intermediate temperatures use R(T) = 10^((37−T)/2), i.e. one order of magnitude per 2 °C: a log-linear interpolation between the two measured points, not a claim of measurements in between. The defense index is the inverse mirror: the papers measured the direction (type I/III interferon and interferon-stimulated genes are induced far more strongly at 37 °C than 33 °C), so the mirror shows direction, and the 100× magnitude belongs to the viral titer.
Human confirmation. The 2015 landmark was mouse cells. The 2016 follow-up (Foxman et al., PNAS) reproduced the temperature dependence in human bronchial epithelial cells and H1-HeLa cells, so the effect is not a mouse artefact, but the headline 100× figure is the mouse number.
Reproduce offline: node research/wet-hair-cold-weather-catch-a-cold/verify-wet-hair-cold-weather-catch-a-cold.mjs
The honest counterweight
There is a genuine tension in the literature, and the honest answer names it. The 1968-era chilling studies found nothing. But in 2005, Johnson & Eccles chilled 180 volunteers' feet in cold water for 20 minutes. Of the 90 chilled, 13 reported a cold in the following days; of the 90 controls, only 5 did, a difference that just reached significance (P = 0.047).
That is not a debunk of layer one; it is the ruler-twist again. Eccles' own reading: a chill triggers reflex constriction of nasal blood vessels, cooling the lining and cutting local defense, which can tip a silent infection you already carry into felt symptoms. It never makes a cold from nothing; it worsens the outcome of a virus already present. Same causal order.
| Johnson & Eccles 2005 | reported a cold | did not | rate |
|---|---|---|---|
| chilled feet (n=90) | 13 | 77 | 14.4% |
| control (n=90) | 5 | 85 | 5.6% |
| ratio / significance | relative risk 2.6× | P = 0.047 | |
Sources: dated, primary, and labelled mouse vs human
2015 · PNAS 112(3):827–832 · pub. Jan 5 2015mouse cells
The landmark. ~100× higher final rhinovirus titer at 33 °C vs 37 °C in wild-type mouse airway cells; type I interferon & ISG induction far stronger at 37 °C. doi:10.1073/pnas.1411030112
2016 · PNAS 113(30):8496–8501human cells
The human confirmation. Same temperature dependence in human bronchial epithelial and H1-HeLa cells; adds apoptosis and RNaseL as extra warm-temperature defenses. doi:10.1073/pnas.1601942113
2005 · Family Practice 22(6):608–613people
Johnson & Eccles. "Acute cooling of the feet and the onset of common cold symptoms."
The contested counterweight. Chilled feet raised self-reported colds 13/90 vs 5/90 (P=0.047), consistent with worsening a latent infection, not creating one. doi:10.1093/fampra/cmi072
1946–1989 · MRC Common Cold Unit, Salisburyhistory
The Common Cold Unit (Andrewes / Tyrrell)
Decades of deliberate chilling (cold walks, wet socks, unheated rooms) failed to give volunteers colds without a virus. "Chilling by itself" did nothing. This is layer one.
What's proven, what's modelled, and what I'm careful not to claim
Proven. Colds are viral (mostly rhinovirus). Chilling with no virus present does not create a cold: the Common Cold Unit's decades of trying. Rhinovirus replicates more at 33 °C than 37 °C, and the airway interferon/ISG response is weaker at 33 °C: the direction is robust across mouse (2015) and human (2016) cells.
Modelled, not measured. The two measured endpoints are 33 °C and 37 °C (the ~100× titer ratio). Every intermediate temperature on the slider and in the table is a log-linear interpolation between them: a smooth trend I drew through two real points, not additional measurements. The defense curve shows the measured direction; its 100× scale is borrowed from the viral titer as a visual mirror, not an independently published ISG fold.
What I will not claim. This is a cell-culture replication and interferon effect. It is not a real-world "cold noses make you X% more likely to get sick" number; no such clean epidemiological figure exists, and I refuse to invent one. Cold exposure is not stated to be "harmless"; the honest claim is narrower: chill modulates the outcome of an infection already present, it does not manufacture one.
Seasonality is over-attributed if you stop here. Colds rise in winter for many reasons at once: indoor crowding, low humidity that dries mucus and stabilises virus, less UV/vitamin D, more time sharing air. Nasal temperature is one plausible strand, not the sole cause of the winter peak.