I am currently comparing technologies that could provide a physical response rather than only notifying someone that their physiological state may have changed. This paper stood out because it investigates affective touch as a wearable, in-the-moment response to anxiety.

The idea is different from a standard vibration alert. Instead of asking the user to interpret a notification or deliberately follow a breathing exercise, the wearable creates a slow stroking sensation on the skin. For my project, this makes it useful for thinking about whether a tactile response could support regulation while someone is already occupied with a demanding task.

What affective touch is

The paper describes affective touch as slow, low-force stroking on hairy skin. My notes focused on the proposed biological mechanism behind this. The authors identify C-tactile (CT) afferents as a primary pathway for affective touch. These nerve fibres respond particularly strongly to gentle stroking at slow speeds, with peak activation at around 3 cm/s.

The paper connects this CT-optimal touch with pleasantness and reward pathways, including the brain's opioid system. I found this interesting because the intervention is not simply using touch as a message. The physical qualities of the touch — softness, force and speed — are part of the proposed mechanism.

This also means that the response cannot be treated as an arbitrary haptic pattern. If I explored affective touch further, I would need to preserve the characteristics that are thought to make the touch affective rather than assuming that any vibration or movement on the skin would have the same effect.

How the wearable works

The prototype was worn on the back of the forearm, where CT afferents are found in hairy skin. A small piece of synthetic fur moved backwards and forwards across the skin using a motor, pulley and spring mechanism. The stroking distance was approximately 3 cm and the device was designed to move at a CT-optimal slow speed.

The prototype measured around 170 × 50 × 25 mm and weighed 77 g. It was held to the arm with Velcro straps. This is important for my design research because the technology is physically understandable: it is essentially a wearable mechanism that repeatedly strokes the forearm with a soft surface.

The researchers prioritised reproducing the validated brush-stroking mechanism rather than creating a highly refined everyday wearable. They explicitly describe the prototype as relatively obtrusive and suggest that future versions would need a smaller and more wearable-friendly form.

What the researchers tested

The study recruited 32 adults, with 24 valid datasets remaining after exclusions. Twelve participants were analysed in the control group and twelve in the experimental group.

Both groups wore the device while completing an adapted Trier Social Stress Test involving speech preparation, unexpected mental arithmetic and a speech task. The experimental group received the stroking sensation during these stressful tasks. The control group wore the same device but did not receive the stroking sensation until the later device try-out.

The researchers measured self-reported state anxiety between tasks and recorded heart activity using a Polar H10 chest strap to calculate heart-rate-variability measures. This gave them both a subjective anxiety measure and physiological data.

What stood out to me

One of the findings I marked was that the experimental group reported significantly lower state anxiety during the mental arithmetic task. The change in state anxiety was also lower across the speech preparation, mental arithmetic and speech tasks.

I also found the sensory descriptions particularly useful. Most participants described the sensation positively, but four participants experienced it as itchy or ticklish, and one described the repeated stroking as odd or unnatural. This is a useful reminder that a sensation designed to be pleasant is not automatically pleasant for everyone.

Twenty-three of the 24 participants described the sensation as calming or soothing, and most reported that it was not very distracting once they became used to it. However, the minority who disliked the sensation matters just as much for my project because tactile preference is likely to be highly individual.

What the results actually show

The paper provides evidence that wearable affective touch can influence reported state anxiety during laboratory stress tasks. The experimental group showed lower anxiety than the control group, with the clearest individual task difference reported during mental arithmetic.

The physiological result was more complicated. The intervention did not produce the expected immediate reduction in the main physiological stress-response measures. The authors therefore distinguish between the reduction in reported anxiety and the lack of a clear immediate physiological calming effect.

This distinction is important for my project. A person feeling less anxious does not necessarily mean every physiological signal will immediately move in the same direction. It reinforces the need to avoid treating physiological data as a simple truth about how someone feels.

The connection to biosensing

The wearable itself was not controlled by live biodata in this experiment. ECG was collected to evaluate the intervention, but the participants' physiological signals did not determine when the stroking started or how the device moved.

What interests me is that the authors explicitly discuss a future just-in-time system in which bodily signals could be monitored and used to decide when affective touch might be offered. They describe different levels of control, from automatic activation through to a consent-based mode and fully manual activation.

For my project, the most plausible role for biodata would therefore be to influence when the response is offered or activated, rather than to continuously change every physical property of the touch. It could potentially contribute to deciding when to start a stroking sequence, how long support remains available, or when to ask the traveller whether they want it to continue.

I would be more cautious about using biodata to continually change the stroking speed. The paper's mechanism depends on touch staying within a CT-optimal range, so freely increasing or decreasing speed could move the sensation away from the characteristics the research is actually testing. This is a useful design constraint: biosensing does not need to control everything simply because the technology allows adaptation.

User control feels central

A particularly relevant part of the paper is that participants wanted more control over when and how they received the touch. Some wanted to initiate it themselves, some wanted to be able to stop it easily, and participants differed in the situations where they thought the intervention would be appropriate.

The authors therefore propose three possible control modes: device-controlled, consent-based and fully user-controlled. This closely connects with my existing project direction. A system could notice a sustained physiological change and offer support, while the traveller still decides whether they actually want the tactile response.

I find the consent-based model particularly interesting as a design possibility. It sits between a purely manual system and one that acts on the body automatically. It could reduce the need for the traveller to continually monitor themselves while still avoiding the assumption that a detected signal gives the system permission to intervene.

Autism and sensory relevance

This study did not involve autistic adults and it was not conducted in an airport. I therefore cannot use it as evidence that affective touch will regulate autistic travellers or reduce sensory overload.

What it does provide is evidence for a specific tactile mechanism in another population and context. For my project, the relevance is therefore a design inference: slow, soft touch may be worth exploring as one possible personalised response for autistic adults who find that kind of tactile input acceptable.

The negative sensory responses in the study make this especially important. If some participants without an autism-specific recruitment criterion already experienced the stroking as itchy, ticklish or unnatural, I should not assume that it would be universally comfortable for autistic users. A future system would need personal preference, intensity limits, immediate stop controls and the ability to choose a completely different response modality.

Limitations for an airport context

The study took place in a controlled laboratory. Participants were seated, wore a wired prototype and completed deliberately induced stress tasks. This is very different from walking through security, carrying luggage, navigating crowds, processing announcements or responding to unexpected changes in an airport.

The prototype itself was also relatively bulky and was not designed as a finished everyday wearable. The researchers note that the brush-based mechanism is obtrusive and suggest exploring smaller alternatives such as haptic illusions.

There is also an important difference between anxiety and the wider sensory, cognitive and situational demands I am investigating. Even if affective touch can reduce state anxiety in some situations, that does not establish that it addresses sensory overload, disorientation or every form of regulation need.

What this means for my project

This paper strengthens my interest in tactile responses because it shows that a wearable can do more than buzz or display information. The physical response itself can be designed around a specific sensory mechanism.

The most useful shift in my thinking is that I can separate what biodata controls from what the tactile mechanism needs to remain effective. If affective touch were explored within a bioresponsive system, physiological information might help determine timing, while the stroking sensation remains within an appropriate physical range and the traveller retains control over whether it happens.

It also makes personalisation more concrete. Personalisation is not only choosing a different colour or interface setting; it may mean deciding whether touch is wanted at all, where it is felt, when it can activate, how long it continues and how easily it can be stopped.

I am keeping affective touch as a promising tactile-response reference rather than a proven airport intervention. The next question is whether a softer, less obtrusive textile or haptic system could reproduce the relevant stroking sensation while remaining comfortable during movement — and whether intended users would actually want that form of support.

Source

Zhao, Y., Tao, Y., Le, G., Maki, R., Adams, A., Lopes, P. and Choudhury, T. (2022). Affective Touch as Immediate and Passive Wearable Intervention. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 6(4), Article 200, pp. 1–23.

Key reading from my notes: CT afferents and affective-touch mechanism, p. 4; state-anxiety findings, p. 11; participant sensory responses, pp. 12–13; control, just-in-time intervention and limitations, pp. 15–18.