mariuscomper.uk Română

A hand on an arm

How slowly a hand has to move

Draw one finger along the arm. This page follows two of the ways the skin answers. Only one of them prefers a slow stroke at the warmth of skin. The speeds were measured in forearm nerves, and in ratings of a brush.

The sound starts when your hand moves, at a quiet level. Boost makes only the slow sound louder, so a phone speaker can carry it. The glass under your finger is not skin.

Nothing is moving yet

An arm, in lamplight. The recordings used here begin with the probe already moving.

A hand in a hurry

30 centimetres a second

A stroke of 10 centimetres, the length of those brush strokes, would take a third of a second. Rochelle Ackerley and colleagues recorded single nerve fibres in the left forearm while a smooth metal probe stroked at five speeds. Four hair fibres were the ones compared with the C-tactile (CT) afferents, slow unmyelinated touch fibres. Their firing rose as the probe sped up. Its temperature did not change them.

The brighter sound rises with speed, in that direction. It is not a spike count. The paper does not print those rates as a table. The lower sound follows a rating curve. Neither is a recording of a nerve.

The other touch

3 centimetres a second, 32 degrees

The average firing of the slow fibres was highest at that speed and at 32 degrees, about the temperature of the skin on an arm. The rate itself is in a figure, not a table. Eight fibres finished that recording. Twenty had been found, conducting at 0.76 metres a second on average.

Thirty other people, in a separate session, rated the same five speeds. For every one of them, a curve with a peak fit better than a straight line. That does not place each person's own peak at 3 centimetres a second. Across five studies pooled together, 127 people still produced that curve as a group, and only 42 percent of them showed a significant curve of that shape for themselves (Croy, Bierling, Sailer and Ackerley, 2021). Across the five speeds, at skin temperature only, the average firing and the average rating rose and fell together. The people and the fibres were different groups. A match of two averages leaves the cause open. At skin temperature, the group curve accounts for about a tenth of the variation in the firing, and about a tenth of the variation in the ratings.

In a separate experiment, Chantal Triscoli, Rochelle Ackerley and Uta Sailer had 12 people rate a goat-hair brush on the forearm. Each stroke ran about 10 centimetres, at a force of 0.4 newtons. On a scale from minus 10 to plus 10, the mean was 4.04 at 3 centimetres a second, minus 0.56 at 0.3, and 0.50 at 30. The middle speed was rated higher than either extreme. The spread between people was about as wide as the gap.

In two smaller groups, one of 8 and one of 9, each stroked at only one speed for about 50 minutes, the difference between 3 and 30 was not significant. The higher rating at 3 showed up when people could compare speeds.

Slower than that is below the peak

0.3 centimetres a second

At a third of a centimetre a second the average firing of the slow fibres is below the peak at 3. They are not described as silent. At that slow speed, skin temperature still produced more firing than a cold probe and more than a hot one.

The hot probe came out lower

The same probe was also set to 18 degrees and to 42. From 0.3 to 10 centimetres a second, skin temperature was rated more pleasant than the cold probe and more pleasant than the hot one.

The fibres agreed, with one exception the paper prints. At exactly 3 centimetres a second, skin temperature fired significantly more than the cold probe. The comparison with 42 degrees was not significant at that single speed. At 30 centimetres a second, temperature did not change the firing.

Choosing a cold or a hot probe lowers the slow sound by six decibels, so the drop can be heard. That step is not a ratio from the paper. The paper reports that the difference was significant.

Now the speed is yours

The brush strokes were about 10 centimetres. The band on the glass is not. Draw along it with one finger. The three buttons play the published speeds, matched or not.

Match the line to 5 centimetres on a ruler, or to the short edge of a bank card.

The short edge used here is 53.98 millimetres, the size of an ID-1 card.

After that match, the number at the bottom is your speed in centimetres a second. Before it, the number is marked as mapped: the whole band stands for those 10 centimetres.

The three buttons are the way through the sounds without drawing. They play 0.3, 3 and 30 centimetres a second.

The low sound is the rating curve: the three means above, joined on a logarithmic speed scale and held still outside them. The bright sound rises with speed, in the direction of the hair fibres. Neither is a recording of a nerve. Both are filtered noise. A finger on glass is not a brush at 0.4 newtons, and force is not measured here.

Pleasantness at three speeds Means on a scale from minus 10 to plus 10. 100-10 0.3330 cm/s
Filled dots are the three means from the 12 people. The line between them is the interpolation the sound uses. The mark follows the speed in the sound.

In a study of 31 people, two identical brushes, one in a hand and one on a robot, did not differ in pleasantness as a main result. A smaller interaction between hand and speed did not survive a correction for several tests. The speed did matter. The hand felt stronger, as intensity, at the two slower speeds.

The palm is a different map

These recordings were from the forearm, where the skin has hair. In 2014 the same group wrote that this kind of fibre had not been found in the palm. With 34 people and the five speeds, pleasantness on the palm did not fall away at the fast end the way it did on the arm.

In 2021, Roger Watkins and colleagues reported three such fibres in the hairless skin of the hand. They estimated them about seven times rarer than in hairy skin, and they left open what those fibres do there. In 2023, Schirmer, Croy and Ackerley concluded that these fibres support gentle affective touch, and that an affective touch need not rely on them or feel pleasant.

When the arm belonged to someone

Ilona Croy and colleagues asked people to stroke, and published the summary in the abstract. All 32 who stroked a partner, and all 11 parents who stroked a baby, used a speed the authors describe as able to activate these fibres. Those speeds were slower than the speeds 45 other people used on an artificial arm.

The full tables of that paper are not open. The sentence is the abstract's, not a reading of each hand.

The arm is yours

The 3 centimetres a second reference is the one coming back. Near 3 centimetres a second, with the probe at skin temperature, both sounds are present. If you were fast, the bright sound is the louder of the two. The slow sound is still the rating for that speed, not silence.

A slow stroke lengthened the gap between heartbeats more than a fast stroke did, in 29 people, and that difference did not depend on forearm versus palm. Ralph Pawling and colleagues read the slowing as the body orienting to a touch, and the palm result, they wrote, counts against explaining it by this nerve alone.

They are closer than the glass

If someone is next to you and wants your touch, try slowing into that range. The plate in the recordings was at the warmth of skin, and that was rated above a cold plate and a hot one.

Then put the phone down.

Sources and method

The low sound follows three published means, from Triscoli, Ackerley and Sailer (2014), Experiment 1: 12 people, the dorsal surface of the left forearm, a goat-hair brush, about 10 centimetres, 0.4 newtons, a scale from minus 10 to plus 10. The means are 4.04 at 3 centimetres a second, minus 0.56 at 0.3, and 0.50 at 30. Between those speeds the sound moves on a logarithmic scale. Outside them it holds the end value. The slow sound drops by three decibels for each rating point below the peak mean of 4.04. The ratings are not decibels.

The temperature step is six decibels quieter for a probe at 18 or 42 degrees, and only at speeds up to 10 centimetres a second. That follows the ratings, which were higher at skin temperature than at either other temperature from 0.3 to 10 centimetres a second. Ackerley, Backlund Wasling, Liljencrantz, Olausson, Johnson and Wessberg (2014) do not print the difference in decibels. For the fibres, at exactly 3 centimetres a second, skin temperature was significantly above the cold probe only. At 30 centimetres a second their fibres showed no temperature effect, so the slow sound does not duck there. The brighter sound runs from minus 16 decibels at 0.3 centimetres a second to 0 at 30, along the same logarithmic speed. That slope is the direction of the hair fibres, not their spike counts. Boost adds nine decibels to the slow sound only. A master level of minus 14 decibels keeps the default quiet. Both layers are filtered noise. There is no heartbeat.

A speed drawn with a finger is in centimetres a second only after the line is matched to 5 centimetres on a ruler, or to the short edge of an ID-1 card, 53.98 millimetres (ISO/IEC 7810). Before that match the band is scaled onto the 10 centimetre stroke and the number is marked as mapped. The buttons at 0.3, 3 and 30 are the published speeds and do not use the match.

Eight fibres completed the temperature protocol. Twenty had been found, conducting at 0.76 metres a second on average. The pleasantness ratings at the five speeds came from 30 other people. A curve with a peak fit better than a straight line for every one of the slow fibres tested that way, and for every one of those 30. The group curve at skin temperature accounted for about a tenth of the variance (R² = 0.10) in the firing and about a tenth in the ratings. The correlation of the five average firing rates with the five average ratings, at skin temperature only, was R² = 0.96. Matching the three means shows that the sound follows that table. It does not make the biology certain.