Nobody teaches you how to stroke someone’s arm. There is no lesson, no diagram, no correct answer passed down. And yet if you filmed a hundred people doing it — a partner’s forearm, a friend’s shoulder, a baby’s back — their hands would move at strikingly similar speeds. Slower than you would guess. Slower than a hand moves for any practical reason.
There is a set of nerve fibres that responds best at almost exactly that speed, and it has been recorded, one axon at a time, in awake human beings.
The claim, stated plainly
The idea in circulation is that we have a dedicated nerve for affectionate touch. That is a compression of something more careful, but the compression is not far wrong, which is unusual for a claim this shareable.
Where it came from
The method is microneurography: a fine electrode into a nerve in a conscious volunteer’s arm, listening to single fibres. In 1999 a Swedish group reported 38 unmyelinated units in the forearm skin. Twenty-seven of them had low thresholds and responded to gentle, harmless deformation of the skin — not to damage. They named them tactile afferents, and argued they constitute a second tactile system in hairy skin, running alongside the fast myelinated one everybody already knew about.1
A second system is a claim about architecture. What it is for took another decade.
The speed
In 2009 the same tradition of work stroked the skin with a soft brush at a range of velocities while recording from single fibres and asking people how pleasant it felt. The C-tactile afferents — and not the myelinated afferents — responded most vigorously at intermediate speeds, between roughly 1 and 10 centimetres per second. Those were the speeds people rated as most pleasant.2
Two curves, one from a nerve and one from a person, with the same inverted-U shape. Slower is not better and faster is not better; there is a band, and the band is narrow. This is the finding the whole field rests on, and it has held up.
The temperature
Then a detail that is easy to miss and hard to forget. In 2014 researchers stroked the skin at five velocities — 0.3, 1, 3, 10 and 30 cm/s — at three temperatures: cool at 18°C, neutral at 32°C, and warm at 42°C. The C-tactile fibres discharged preferentially to the slow speeds at neutral skin temperature. Not the cool stimulus. Not the warm one.3
Thirty-two degrees is the temperature of ordinary skin. These fibres are, on this evidence, tuned to the speed and the warmth of another person’s hand. That is a small number of experiments and it is a brush rather than a hand, so hold it as a tuning curve rather than a law. But it is the sort of finding that reorganises how you read your own behaviour.
Out of the lab
The obvious next question is whether anyone actually touches like that when nobody is measuring. In 2016 a group ran three experiments to find out: 45 people stroking an artificial arm, 32 stroking their partner, and 11 parents stroking their babies. Spontaneous stroking landed in the C-tactile-optimal band.4
Nobody had been told the number. The behaviour arrives before the explanation, which is roughly always the order these things happen in.
The patient who could not feel her own hands
The most striking evidence is a single person. G.L. had lost her large myelinated afferents, the fast fibres that carry ordinary discriminative touch. She could not feel a poke, a texture, a shape. When researchers brushed her hairy skin gently, she reported a faint but distinctly pleasant sensation — and imaging during that stimulation showed activity in the insular region rather than the usual somatosensory areas.5
A person with the touch system removed could still receive the affectionate part. That is one patient, described in detail, and single cases are evidence of possibility rather than of rate. But it is very hard to explain any other way.
Where it gets complicated
Here is the part that has moved since, and the reason this piece is a Solid with an asterisk rather than a triumph.
The tidy version — C fibres do feeling, fast fibres do information — was never quite what the original authors argued,6 and it does not survive testing. In 2023, when A-fibre function was greatly diminished by nerve block, the perceived pleasantness of exactly the sort of slow stroking these fibres prefer was nearly abolished — and the same happened for deep pressure.7
So the explicit experience of “that was lovely” appears to need the fast fibres too. The C-tactile system looks less like a private channel for affection and more like one necessary voice in a chord. The speed tuning is real and replicated. The story about what it delivers on its own is still being argued.
One thing worth sitting with
Three centimetres a second is slower than almost anything else your hand does. It is slower than reaching, slower than wiping a surface, far slower than a reassuring pat. Somewhere in your arm are fibres that answer to that speed and largely ignore the rest — and when you comfort someone without thinking, that is roughly the speed you choose.
You have nerves that answer to a slow stroke and mostly ignore a fast one — tuned to about the speed, and about the warmth, of an unhurried hand. Nobody teaches it. Filmed doing it unprompted, people land in the band anyway.
What this might mean for an ordinary evening
- Slower than feels natural. If you are consciously trying, most people go too fast. The band tops out around 10 cm/s and the sweet spot sits nearer 3.
- Hairy skin, not palms. These fibres were found in forearm skin and are absent from the glabrous skin of the palm. Forearm, shoulder, the back of the neck.
- Warm hands. The tuning is to ordinary skin temperature. Cold hands are not merely unpleasant; on this evidence they are off-band.
- Don’t perform it. The behaviour is already correct when it is unselfconscious — and monitoring your own technique is its own way of leaving the room, which we wrote about in Spectatoring.
We wrote separately about what touch may be doing further downstream, and how much of the oxytocin story survives the papers, in Touch, Closeness, and the Quiet Hormone. Our closeness work is company for an unhurried hour; it claims nothing described above.
References
- Vallbo, Å. B., Olausson, H. & Wessberg, J. (1999). Unmyelinated afferents constitute a second system coding tactile stimuli of the human hairy skin. Journal of Neurophysiology, 81(6), 2753–2763. doi:10.1152/jn.1999.81.6.2753
- Löken, L. S., Wessberg, J., Morrison, I., McGlone, F. & Olausson, H. (2009). Coding of pleasant touch by unmyelinated afferents in humans. Nature Neuroscience, 12(5), 547–548. doi:10.1038/nn.2312
- Ackerley, R., Backlund Wasling, H., Liljencrantz, J., Olausson, H., Johnson, R. D. & Wessberg, J. (2014). Human C-tactile afferents are tuned to the temperature of a skin-stroking caress. Journal of Neuroscience, 34(8), 2879–2883. doi:10.1523/JNEUROSCI.2847-13.2014
- Croy, I., Luong, A., Triscoli, C., Hofmann, E., Olausson, H. & Sailer, U. (2016). Interpersonal stroking touch is targeted to C tactile afferent activation. Behavioural Brain Research, 297, 37–40. doi:10.1016/j.bbr.2015.09.038
- Olausson, H., Lamarre, Y., Backlund, H., Morin, C., Wallin, B. G., Starck, G., Ekholm, S., Strigo, I., Worsley, K., Vallbo, Å. B. & Bushnell, M. C. (2002). Unmyelinated tactile afferents signal touch and project to insular cortex. Nature Neuroscience, 5(9), 900–904. doi:10.1038/nn896
- McGlone, F., Wessberg, J. & Olausson, H. (2014). Discriminative and affective touch: sensing and feeling. Neuron, 82(4), 737–755. doi:10.1016/j.neuron.2014.05.001
- Case, L. K., Madian, N., McCall, M. V., Bradson, M. L., Liljencrantz, J., Goldstein, B., Alasha, V. J. & Zimmerman, M. S. (2023). Aβ-CT affective touch: touch pleasantness ratings for gentle stroking and deep pressure exhibit dependence on A-fibers. eNeuro, 10(5), ENEURO.0504-22.2023. doi:10.1523/ENEURO.0504-22.2023