I am trying to understand what a bioresponsive wearable could actually do back to the body once it has detected a physiological change. This paper was useful because the sensing and the response are directly connected: heart rate is measured in real time and used to control a vibration pattern on a smartwatch.

Unlike a simple notification, the vibration is intended to act as a passive relaxation cue. The user does not have to look at a screen, interpret a graph or deliberately match a breathing exercise. That makes it relevant to my interest in support that could operate quietly in the background while someone is already managing a demanding environment.

What the technology is

The researchers developed a custom Apple Watch system that delivers rhythmic haptic biofeedback through the watch’s Taptic Engine. The watch measures the wearer’s current heart rate and continually adjusts the timing of the vibrations.

The important part for me is that the feedback is closed-loop. Rather than playing one fixed rhythm, the system updates its vibration frequency from the person’s live heart rate. The vibration tempo was set approximately 4% slower than the current heart rate and recalculated every second.

The intention is a form of tactile entrainment: giving the body an external rhythm that is slightly slower than its current cardiac rhythm, with the aim of encouraging reduced arousal.

Why the rhythm matters

The researchers did not only test one repeated buzz. They created five rhythmic patterns informed by lullaby structures, including simple taps, longer vibrations, alternating patterns and two versions of a 3/4 rhythm.

I found this useful because it shows that the pattern of haptic feedback matters as well as whether vibration is present. The more varied 3/4 and alternating rhythms were experienced as more relaxing than the monotonous tap and vibration patterns.

This moves haptics away from the idea of a smartwatch simply “buzzing when something is wrong”. If I use haptic feedback in my project, the pacing, rhythm, repetition and sensory quality would all need to be designed deliberately.

What the researchers tested

The paper contains two studies with healthy adult participants.

In the first study, 20 participants experienced five adaptive vibration patterns while seated in a controlled resting context. Each pattern was delivered for five minutes, with heart rate and subjective ratings collected.

The second study involved 28 participants and took the most preferred 3/4 rhythm into a longer 20-minute sleep-onset condition. Participants completed both vibration and no-vibration conditions, while heart rate, heart-rate variability, perceived arousal, sleep-onset measures and psychomotor performance were recorded.

This is important for interpreting the results because the paper is primarily about relaxation and sleep initiation, not anxiety treatment, autism or use while moving through a public environment.

What stood out to me

One of my strongest notes was that all five vibration patterns significantly reduced heart rate compared with baseline during the short resting study. The patterns did not significantly differ from one another in their heart-rate effect, even though participants clearly preferred some rhythms over others.

That separation between physiology and experience is interesting. A pattern can produce a similar heart-rate change while feeling more or less relaxing, pleasant or intrusive to the person wearing it.

The 3/4 rhythm was the strongest preference overall. Half of the participants ranked it first, and the 3/4 and alternating patterns were rated as more relaxing than the simpler repetitive patterns.

What the results actually show

The short study provides evidence that adaptive rhythmic vibration was associated with acute heart-rate reduction in this controlled resting context.

The longer study was much less conclusive. The researchers found trends in HRV measures that were consistent with increased parasympathetic activity, but these differences were not statistically significant. There were also no significant differences between vibration and control conditions for heart rate, perceived arousal, sleep-onset latency or psychomotor performance.

This makes the paper more useful to me than a simple positive headline. The intervention appeared promising for short-term relaxation, but the effect did not translate clearly into improved sleep initiation during the longer condition.

One participant also reported that the vibration eventually seemed to keep them awake. This raises the possibility that a response which is useful briefly could become distracting or overstimulating if it continues for too long.

The connection to biosensing

This paper gives me a very concrete answer to the question “what could biodata actually control?”

Here, heart rate controls the tempo of the haptic response. The system continually recalculates the vibration timing so that it remains slightly slower than the user’s current heart rate.

For my project, that suggests several possible variables that physiological data could influence: when a haptic sequence is offered, its pace, how long it continues, or whether the system gradually reduces stimulation as the person’s signals change.

However, the paper also exposes an important limitation. The researchers did not directly compare the adaptive closed-loop rhythm with an equivalent fixed-frequency rhythm. They therefore cannot establish whether the bioresponsive element itself caused the effect, or whether the rhythm and tempo would have produced similar relaxation without live heart-rate sensing.

That directly relates to one of my current project questions: does biosensing actually add value compared with a simpler manually selected or fixed response?

Sensory experience and wearability

The use of an ordinary smartwatch makes the technology immediately understandable and relatively discreet. It does not require a specialist headset or large piece of equipment, and the feedback can happen without visual attention.

At the same time, the paper shows that familiarity with smartwatch vibration can change the experience. Some participants associated the haptic sensation with notifications or alarms, making it more alerting or distracting rather than calming.

The authors also suggest that direct wrist vibration may not be the ideal physical form. They discuss alternative body locations and softer forms of tactile actuation as possible ways to make the response less intrusive.

This is relevant to my own move towards materiality and sensory comfort. A technically effective signal can still fail if its physical sensation has the wrong meaning for the wearer.

Autism and airport relevance

This study did not involve autistic participants and it did not test the system during travel. I therefore cannot use it as evidence that rhythmic haptic biofeedback will regulate autistic adults in airports.

The mechanism is still relevant as a design reference. Haptic feedback is non-visual and can potentially operate without blocking environmental sound, which could be useful where a traveller needs to keep attending to announcements, signs and movement around them.

However, the airport context introduces a major complication that the study did not test. Heart rate changes during walking, rushing, carrying luggage, navigating security or responding to unexpected information. A system that simply slows its vibration relative to heart rate could react to ordinary physical activity rather than a need for support.

This means that the closed-loop logic cannot be transferred directly from a resting study into an airport wearable without considering movement, context and personal baseline.

What this means for my project

This paper strengthens haptic feedback as a possible response modality because it demonstrates a real closed-loop relationship between physiological sensing and a tactile output.

The most useful design idea is not necessarily the exact 4% heart-rate reduction. It is the architecture: physiological data can change a physical response in real time, while the response itself can remain subtle and screen-free.

At the same time, the paper challenges me not to assume that “adaptive” automatically means “better”. The study leaves open whether live heart-rate adaptation adds more value than a well-designed fixed rhythm. That is something I could potentially test at prototype level without making any medical claim.

If I take this direction forward, I would want the traveller to retain control over whether a haptic sequence begins, how strong it feels and when it stops. I would also want to compare short, optional sequences against continuous stimulation, because the longer condition in this paper suggests that more feedback is not necessarily more helpful.

I am keeping this as a strong reference for closed-loop haptic biofeedback, but not as proof that heart-rate-driven vibration will regulate autistic airport travellers.

Source

Lee, J., Moschina, D., Ramesh, S., Röddiger, T., Kunze, K. and Beigl, M. (2025). Closed-Loop Rhythmic Haptic Biofeedback via Smartwatch for Relaxation and Sleep Onset. In Proceedings of the 2025 ACM International Symposium on Wearable Computers (ISWC ’25), pp. 30–37.

Key reading from my notes: closed-loop haptic biofeedback and real-time physiological modulation, p. 1; vibration-pattern design and heart-rate adaptation, pp. 2–3; short-term heart-rate and relaxation findings, p. 3; longer sleep-onset results, pp. 4–5; discussion of duration, distraction and the unresolved value of adaptivity, pp. 5–6.