This paper is one of the most directly useful interaction precedents I have read so far because it separates detection from interpretation.

Commercial wearables increasingly generate alerts when they detect physiological changes that may be associated with stress. Neupane et al. ask what happens if the wearable does not stop at detection, but instead uses that moment to invite the person to confirm what they are experiencing and identify the likely stressor.

That is very close to the interaction I am now considering for the airport system.

The most important principle I take from this paper is:

The sensor can decide when it might be useful to ask, but the person decides what the physiological change actually means.

What the study actually did

The researchers developed a system consisting of a smartwatch app, smartphone app and cloud service and tested it in a 100-day field study with 122 participants.

Participants received prompts several times a day when the system detected physiological responses that might indicate stress. Instead of automatically labelling the event, the prompt asked the participant whether they were stressed and, if so, to log the most likely stressor.

Each week, participants were also given visualisations of their own data so they could reflect on patterns and trends.

Across the study, the researchers collected 26,521 momentary stress ratings and 1,057 weekly surveys. The abstract reports reductions over 14 weeks in both the intensity and frequency of self-reported stress, alongside better awareness of stressors and more self-initiated behaviour changes.

This was not simply passive sensing. The system created a loop between physiological detection, user confirmation, reflection and later behaviour.

Detection is not the same as meaning

This is the part of the paper that connects most strongly to my project.

The wearable detects a physiological event, but the system still asks:

“Are you stressed?”

The participant provides the meaning.

That is fundamentally different from a system that says:

“You are stressed.”

For my airport proposition, the equivalent would be:

physiological change detected + context suggests this may be a useful moment → ask the traveller

rather than:

physiological change detected → system decides the traveller is overloaded

Design principle

The sensor can identify a moment worth checking, but the traveller remains the final interpreter of what that moment means.

This is one of the clearest precedents I have found for the kind of relationship I want between biosensing and user control.

What it DOES establish

This study provides good evidence that it is feasible to use wearable physiological events to trigger user-confirmation prompts in everyday life.

That matters because my project does not need biosensing to diagnose sensory overload in order for it to have a useful role.

A more defensible role is:

notice change → create an opportunity to check in → let the person decide

The paper also demonstrates that this interaction can operate over a long field deployment rather than only in a short laboratory session.

Reflecting on data may be as important as the live alert

Another part I found useful is the weekly reflection element.

Participants were shown visualisations of their own patterns and trends rather than only receiving real-time prompts.

This suggests that wearable support does not have to be limited to in-the-moment intervention.

For my system, there could potentially be two different layers:

1. During the airport journey — a lightweight check-in when a meaningful change is detected.
2. Afterwards or between journeys — reflection on where prompts happened, what support was chosen and whether particular environments or stages repeatedly created difficulty.

That second layer could help the traveller configure the system more personally over time.

It could also help avoid the idea that the device itself is the expert. The wearable provides data, but the person gradually builds their own understanding of what patterns matter.

Do not necessarily interrupt immediately

The paper supports event-triggered prompting, but for my airport system I think timing still needs to be treated as a design problem rather than an automatic rule.

A physiological event might happen while somebody is:

• speaking to security staff;
• presenting a passport;
• moving through a scanner;
• boarding;
• finding a gate;
• or dealing with an unexpected announcement.

Even if the sensing is technically correct, interrupting at that exact second may make the situation worse.

So my design interpretation is not:

signal changes → instant alert

It is closer to:

signal changes → system notices → context checks whether this is an appropriate moment → transparent prompt when useful

Interaction caution

A useful trigger is not automatically a useful interruption. The system still needs to consider journey stage, task demand and whether the traveller is in a position to respond.

This is especially important in an airport because situational awareness and completing the immediate task may be more important than responding to the wearable.

The prompt should remain simple

The paper's interaction also makes me think about the wording of my own prompt.

The system does not need to explain a complex physiological model to the user in the moment.

It needs to communicate that something changed, without claiming certainty about why.

For example:

“A change was detected while this area became busier. Would support help?”

The traveller could then choose:

• yes;
• not now;
• dismiss;
• pause prompts;
• or access support manually.

This feels much closer to the evidence than a message such as:

“Your stress is increasing.”

User confirmation also creates better data

There is another useful implication.

When the person confirms whether the prompt was meaningful, that response becomes information the system can learn from.

Over time, the system might begin to distinguish between:

• physiological changes that usually matter to this traveller;
• changes caused mainly by walking or physical activity;
• contexts where prompts are often dismissed;
• and situations where a particular kind of support is repeatedly chosen.

This does not mean the system eventually becomes certain about the person's internal state.

It means that user confirmation can help make future prompting more personalised and less intrusive.

That fits well with the personal-baseline direction emerging from the other papers I have read.

What the paper does not establish

This study does not show:

• that wearable physiology directly detects stress with certainty;
• that the same interaction will work for autistic adults;
• that physiological events identify sensory overload;
• that airport users would welcome repeated prompts;
• that reduced self-reported stress was caused only by the wearable prompts;
• or that a prompt should always appear immediately after a physiological change.

The study is about stress-management behaviour in a general participant sample, not airport accessibility.

The value for my project is the interaction model, not direct evidence for my target population.

What changed in my thinking

Before reading this paper, I was already moving toward a transparent check-in rather than automatic intervention.

This study makes that interaction feel much more defensible because it shows a real field system where physiological detection is used to create a question rather than an answer.

The strongest takeaway for me is:

Biosensing does not need to know what the person feels in order to be useful. It may only need to identify a moment when asking becomes worthwhile.

That is a much better fit with the way I want the airport system to work.

I am also now thinking more carefully about when not to prompt. A good system may need to recognise not only when physiological change occurs, but whether the current airport task makes interruption appropriate.

Source

Neupane, S., Saha, M., Ali, N., Hnat, T., Samiei, S.A., Nandugudi, A., Almeida, D.M. and Kumar, S. (2024). Momentary Stressor Logging and Reflective Visualizations: Implications for Stress Management with Wearables. Proceedings of the CHI Conference on Human Factors in Computing Systems (CHI '24).

Key points from my reading: smartwatch-triggered physiological prompts; 100-day field study; 122 participants; user confirmation of whether stress was present; stressor logging; weekly reflective visualisations; 26,521 momentary ratings and 1,057 weekly surveys; reductions in self-reported stress intensity and frequency over 14 weeks; direct relevance to a sensor asks, person interprets interaction model.