READ / RESEARCH · SOFT TEXTILES AND PROGRAMMABLE TOUCH

Haptiknit is a knitted sleeve containing small air-powered components that press against the skin. By inflating them in different sequences, it can create touch patterns such as taps and strokes. The study explores how clearly people can recognise these sensations and how comfortable the sleeve feels. It offers a useful example of wearable pressure technology, but does not demonstrate reduced anxiety or sensory overload.

Why I looked at this

My sticky note highlights the combination of soft knitted textiles, pneumatic actuators and programmable haptic feedback. This interests me because a wearable response could be integrated into fabric rather than relying entirely on a rigid electronic object.

For my project, I am exploring how touch could support a person during an airport journey. This paper helps me understand how a textile can deliver controlled sensations and what still needs testing before those sensations can be considered helpful.

How the technology works

A pneumatic actuator is a component that moves when air is pumped into it. Haptiknit places eight of these components in a knitted forearm sleeve. As each one inflates, it presses into the skin; different timings and sequences produce different patterns.

The knit combines flexible areas with stiffer regions. This matters because a completely stretchy fabric could expand outwards as the actuator inflates. The stiffer backing helps direct the force towards the skin, while the softer areas allow the sleeve to fit around the arm.

An adjustable fastening helps secure the sleeve. A separate portable pneumatic supply, worn on the upper arm, provides inflation and deflation. The paper reports that this supply system weighs 440 g (du Pasquier et al., 2024, p. 5). The soft textile is therefore one part of a larger wearable assembly.

What the study found

The researchers tested the sleeve with 32 participants. They investigated whether people could identify which actuator was active, how continuous and pleasant stroking patterns felt, and whether touch patterns communicated intended social meanings.

Participants identified the active actuator with 69% accuracy across eight possible locations. This suggests that localised pressure can communicate spatial information, although it was not consistently recognised.

Recognition of the six intended social touch gestures was lower, at 36%. The patterns intended to communicate “calming” and “love” were recognised at approximately chance levels (p. 6). This is an important limit: naming a pattern “calming” does not establish that it makes someone calmer.

The sleeve received median ratings of 8 out of 10 for comfort and 9 out of 10 for ease of putting it on (p. 7). These are encouraging usability findings within the study conditions.

A touch pattern can be noticeable, pleasant or recognisable without reducing distress. Those outcomes need to be tested separately.

Personalisation and control

Before testing, the researchers calibrated the sleeve for each participant. They established when the pressure first became noticeable and when it started to feel uncomfortable, then set an individual maximum for the experiment (pp. 10–11).

This is a useful principle for my project. A standard pressure setting would not account for differences in sensitivity, fit or preference. The wearer should be able to choose an intensity that feels comfortable and stop the response easily.

The study’s patterns were controlled by the experimenter. It does not demonstrate physiological sensing or an automatic response to stress. Connecting the textile to biosensing would be an additional design proposal.

What this could mean for my project

The technology could potentially support a discreet tactile cue or a chosen pressure pattern. Those would serve different purposes: a cue communicates information, while a sustained or repeating sensation might be explored as sensory support.

I should also avoid equating local actuator pressure with a whole-arm squeeze or a weighted garment. Haptiknit’s demonstrated contact patterns are specific to its construction and placement.

A possible future interaction would offer a familiar, user-selected pattern after a meaningful physiological change. The traveller would choose whether to activate it. I would need to compare this with a simple manual control to establish whether biosensing actually improves the timing or usefulness of the response.

Limitations for airport use

Participants reported no cognitive or sensory impairments. They completed seated laboratory tasks wearing noise-cancelling headphones playing white noise. The study therefore does not establish acceptability for autistic adults, performance during movement, or the experience of wearing the system in a busy airport.

Pump sound, weight, warmth, fabric feel, fit under clothing and freedom of movement would all need investigation. The positive comfort ratings are a starting point, rather than evidence of comfortable all-day use.

What I am taking forward

I am keeping Haptiknit as a technical precedent for programmable touch through textiles. Its strongest contribution to my thinking is the combination of soft construction, directed pressure and individual calibration.

The next question is which sensation, if any, an autistic traveller would choose during a demanding moment. That needs to guide the interaction and physical form.

The textile can deliver a pattern. The wearer’s preferences and experience must determine whether that pattern belongs in a support system.

Source

du Pasquier, C. et al. (2024). ‘Haptiknit: Distributed stiffness knitting for wearable haptics’. Science Robotics, 9, article eado3887.

Key reading: sleeve construction, pp. 3–4; portable system and user study, p. 5; gesture recognition, p. 6; comfort ratings, p. 7; participant details and calibration, pp. 10–11. Page references use the article’s printed numbering.