I am currently exploring tactile and pressure-based responses that could provide a more physical, grounding form of support than a screen notification or spoken instruction. This paper stood out because it investigates whether a soft object that appears to “breathe” can reduce state anxiety simply through touch.

My sticky notes describe it as a potentially more natural modality for an anxiety-calming device, which is what interested me most. The intervention does not ask the person to watch an animation, listen to instructions or consciously follow a breathing exercise. Instead, the rhythm is communicated physically through something they hold.

For my project, this gives me another way of thinking about responsive technology: the response itself could be embodied in pressure, expansion and rhythm, rather than always appearing as information.

What the technology is

The final prototype is a soft, huggable cushion containing a pneumatic chamber. An external pump repeatedly inflates and deflates the chamber so that the cushion expands and contracts against the person's body, creating the sensation of slow breathing.

The researchers designed it to sit against the belly and lower chest while being hugged. Its breathing rate was set at 10 breaths per minute, with equal inflation and deflation periods. The intention was to approximate slow breathing and make the movement easy to feel through the body.

What I find useful here is that the output is not vibration. The user feels a gradual change in volume and pressure as the cushion “inhales” and “exhales”. This makes the interaction closer to holding another body that is breathing than receiving a conventional haptic alert.

How the design was developed

Before creating the final cushion, the researchers developed five smaller prototypes with different behaviours inspired by familiar calming sensations, including breathing, purring, heartbeat and illumination.

A focus group with 24 students interacted with the prototypes and rated how pleasant they were. The breathing cushion was rated most positively, with no participant rating it as unpleasant. Participants frequently described it as calming or soothing and said that it reminded them of a person breathing.

This early design stage matters to me because the researchers did not simply choose a tactile effect based on technical novelty. They compared several possible sensory behaviours and let participant response shape which mechanism was developed further.

It also reinforces something I need to keep doing in my own project: the physical sensation has to be tested, not assumed to be calming because the designer intends it to be.

What the researchers tested

The main experiment involved 129 participants, primarily university students. Participants with a clinically diagnosed anxiety disorder were not included in the analysed sample, so this should not be read as a clinical anxiety-treatment trial.

The participants were divided between three conditions: a breathing-cushion condition, a guided breathing meditation and a no-intervention control.

The researchers deliberately created anticipatory anxiety by telling participants that they would shortly have to complete difficult verbal mathematics questions in front of other participants. Before the test, participants spent just over eight minutes either holding the breathing cushion, listening to the meditation or sitting quietly.

Anxiety was measured before and after the intervention, and again after the mathematics test and a second intervention period.

What stood out to me

The clearest result was that, while participants were anticipating the stressful test, anxiety scores were significantly lower in the cushion group than in the no-intervention control group.

The guided meditation also reduced anxiety compared with the control, and the researchers did not find a meaningful difference between the cushion and meditation conditions. In that specific anticipatory phase, the tactile intervention performed similarly to the guided breathing meditation.

That is interesting to me because the cushion gave participants no spoken instructions and no explicit task. They were simply told to hold it.

This makes the intervention feel different from many regulation technologies I have looked at. It offers a physical rhythm that the person can respond to without having to continually interpret instructions.

The result is narrower than the headline

The effect was not equally clear across every part of the experiment.

After participants completed the mathematics test, the second intervention period did not produce significant differences in state anxiety between the cushion, meditation and control groups. Anxiety reduced across the groups after the stressor had ended.

The evidence therefore supports a more specific conclusion: the breathing cushion reduced anticipatory state anxiety in this laboratory test situation compared with no intervention. It does not establish a universal calming effect across all forms or stages of anxiety.

This distinction matters for my project because an airport journey contains very different states: anticipation, waiting, rushing, sensory demand, uncertainty, orientation and recovery. A technology that helps during one type of stress may not automatically help in all of them.

What participants actually felt

The qualitative feedback is useful because the interface was not universally interpreted in exactly the same way.

Most comments about the breathing behaviour were positive. Thirty-three participants described the cushion using terms such as calming, relaxing, comforting or pleasant.

However, a small number found the breathing strange or artificial. Four participants felt that the breathing was too deep or slow, three found it difficult to follow with their own breathing, and one participant became bored over time.

I find this particularly relevant to an autism-focused project. Even within a non-autistic university sample, people differed in how they experienced the same tactile rhythm.

That means I should not treat a “calming” pressure or breathing pattern as a universal sensory setting. Rate, depth, pressure, location and duration would all need to be adjustable or optional.

What may actually be causing the effect

One of the most important limitations is that the study cannot isolate exactly why the cushion reduced anxiety.

The researchers did not compare the breathing cushion with an identical static cushion. This means the observed effect could have come partly from hugging something soft and comforting, from novelty, from the breathing movement, or from some combination of these.

The paper also suggests that the cushion may encourage the person to slow their own breathing through physical entrainment, but the study did not measure participants' breathing closely enough to establish that this mechanism actually occurred.

This is a useful warning for my project. If a prototype appears calming, I need to distinguish between the material comfort, physical pressure, rhythm, expectation and adaptive behaviour rather than attributing the outcome to whichever technological feature sounds most interesting.

Connection to biosensing

This prototype was not bioresponsive. It did not sense the user's physiology and it did not change its breathing behaviour in response to their body.

In fact, the breathing rate stayed fixed throughout the intervention.

The authors suggest that future studies could collect physiological signals such as heart-rate variability, breathing rate and skin conductance to understand the interface's physiological effects and whether breath entrainment occurs.

For my project, this creates a possible design inference rather than an evidence-based conclusion. Biodata could potentially influence when a tactile breathing response is offered, how long it continues, or perhaps the pacing of the expansion and contraction.

However, I would not assume that continuously matching the cushion to physiological data would improve it. The paper tested a fixed slow rhythm, not an adaptive one. Changing the breathing rate in response to arousal could also make the interaction less predictable or move away from the slow-breathing mechanism the design is based on.

A safer design question for me is whether sensing could help decide when to offer a pre-set preferred tactile rhythm, while the person retains control over whether it begins and when it stops.

A possible relationship to pressure and grounding

This paper also makes me think differently about pressure.

The cushion is not a deep-pressure device in the same sense as a compression vest. The contact comes from holding the object and feeling its pneumatic expansion against the torso.

But the combination of softness, contact, slow pressure change and repetitive bodily rhythm suggests a more nuanced tactile territory than a simple squeeze.

That feels useful for my exploration of a “warm hug” or grounding response. A future intervention could potentially use gradual inflation or textile pressure to create a paced tactile experience without copying this cushion form directly.

That is a design possibility, not something the paper proves for autistic adults.

Autism and airport relevance

This study did not involve autistic adults and did not take place in an airport or during travel.

It therefore cannot establish that a breathing cushion, pneumatic pressure or tactile pacing would support autistic travellers experiencing sensory overload.

Its value to my project is as evidence that a specific form of slow, body-based tactile interaction can reduce anticipatory state anxiety in another population and context.

The transfer to airport accessibility remains uncertain. A full cushion is also impractical for someone walking through an airport, carrying luggage or moving through security. The final prototype was powered by an external pump, so it was not an untethered portable device.

For my project, the relevant question is therefore not “should I give travellers this cushion?” but could the underlying tactile mechanism be translated into a wearable or travel-compatible form?

Wearability and sensory burden

The physical form is one of the clearest limitations for my design context.

The researchers deliberately created something soft, ergonomic and suitable for hugging in the home, but the system depended on a mechanical pump and was not designed for mobile use.

A wearable translation would need to consider where the inflation happens, how much pressure is created, how noisy the mechanism is, whether it restricts movement and whether the material feels acceptable against the body.

The paper also raises an interesting issue around simulated human behaviour. A breathing object can feel comforting, but if it feels too artificial it can become strange or distracting.

This reinforces that the emotional meaning of a tactile response is part of the interaction design. A device intended to ground someone cannot rely only on technically reproducing a movement.

What this means for my project

This paper strengthens my interest in pneumatic, pressure-based and rhythmic tactile responses because it demonstrates that haptics can work through gradual bodily movement, not only vibration.

The most useful design idea for me is the combination of a simple tactile rhythm with very low interaction demand. The participant does not need to read, listen or make repeated decisions while the response is happening.

At the same time, the paper makes the case for personalisation. Some participants found the fixed rhythm too slow, too deep or difficult to follow. In my project, this suggests that a tactile intervention would need to be configured around the individual's preferred intensity, rhythm and duration rather than assuming one calming pattern.

If this direction became bioresponsive, I would want sensing to remain secondary to user preference and manual control. Physiological information might help indicate that support could be useful, but the traveller should still decide whether they want the response.

I am keeping this paper as a strong reference for a soft, embodied tactile intervention and the possibility of translating slow breathing into physical pressure, while treating the airport and autism connection as a design question that still needs evidence.

Source

Haynes, A. C., Lywood, A., Crowe, E. M., Fielding, J. L., Rossiter, J. M. and Kent, C. (2022). A calming hug: Design and validation of a tactile aid to ease anxiety. PLOS ONE, 17(3), e0259838.

Key reading from my notes and analysis: rationale for tactile anxiety support, pp. 1–3; initial breathing/purring/heartbeat prototypes and focus-group selection, pp. 5–7; final pneumatic breathing-cushion design, pp. 7–9; anxiety experiment and results, pp. 10–18; participant responses, limitations and future physiological sensing, pp. 19–22.