ClarityBand.
UX Research · UX/UI Design · Product Design · Physical & Digital Design · Prototyping
Sense. Understand. Act.
ClarityBandreadsbrainactivitythroughasoftEEGheadband,turnsitintoasimplebrainstate,andrespondswiththerightkindofresetattherightmoment.HereisthefinalsystembeforethestoryofhowIgotthere.
Inspiration


Why Focus Needs a Different Approach
I explored what disrupts attention and how existing focus-support strategies address these challenges, looking for opportunities beyond simply blocking distractions.
Sustained Focus Is Short, and It Isn't the Same for Everyone
Before blaming myself, I wanted to know whether short focus was a personal problem. Secondary research showed it is common, and that how long people can stay focused shifts with where they live and how old they are.
Sustained attention varies across the lifespan, with differences in performance between children, young adults, and older adults.
Distraction Comes from More Than One Source
Once I knew short focus was common, I wanted to hear how it plays out day to day: what pulls people away, whether they notice it happening, and what they do to get back on track.
Participants discussed different causes of distraction and described moments when their attention shifted away from tasks. We worked through four questions and clustered the answers on a shared board.
The discussion revealed three patterns:
Existing Solutions Address Only Parts of the Problem
Findings 01 and 02 gave me four things a focus tool would need to handle: different personal rhythms, more than one source of distraction, drift that is noticed too late, and recovery that runs on willpower. I used them as a lens to look at what people already rely on.
I compared four representative products and the most common self-discipline advice against those four needs. I chose one well-known physical product for each of the four ways people try to manage focus: blocking noise (Bose QuietComfort), masking the environment (LectroFan), planning ahead (Full Focus Planner) and sensing attention (Foci). I left out phone apps, since the phone is itself one of the biggest distractions; screen-time tools are covered under "Common advice".
| What a focus tool needs | Bose QC | Full Focus Planner | LectroFan | Foci | Common advice |
|---|---|---|---|---|---|
Adapts to the individual From Finding 01 | |||||
Covers more than one source From Finding 02 | |||||
Notices drift in the moment From Finding 02 | |||||
Works without willpower From Finding 02 |
Mapping the same options by when they act and how much effort they ask for shows where the gap is.
Most options either block a single source of distraction or rely on willpower set up in advance. Foci is the only one that responds in the moment, but it depends on the phone and takes effort to learn. Nothing yet notices drift as it happens while staying effortless to live with.
From Managing Distractions to Supporting Attention
People experience attention differently, distractions come from multiple sources, and existing strategies often require ongoing effort.
Can a brain signal show distraction?
The research helped me understand how people experience distraction. Next, I compared a distraction-free and a distracted version of the same kind of task, to explore how task performance and EEG readings change when attention is pulled away.
+Theoretical basis: why EEG?· Kaushik et al., 2022
Research has shown that EEG activity is significantly related to attention levels. In focused states, the brain, especially the frontal cortex, shows more high-frequency activity (beta and gamma waves), indicating intense cognitive processing. In unfocused or relaxed states, alpha activity increases, particularly in the occipital and parietal regions, reflecting a relaxed or resting brain.
Kaushik, P., Moye, A., Vugt, M. van and Roy, P.P. (2022). Decoding the cognitive states of attention and distraction in a real-life setting using EEG. Scientific Reports, 12(1). doi:10.1038/s41598-022-24417-w
Setup
What changed
The readings did shift when distractions appeared, which made me think about how to use them. But a change in the signal doesn't tell the system when to step in. That raised the questions the next part had to answer:
From a signal to the right moment.
The experiment showed that the readings move. It didn't say when a device should act on them. Each question below started with an obvious answer that broke down in a real moment of work.
Q1
My first instinct was simple: when the reading crosses a threshold, buzz.
But a dip isn't always distraction. Picture someone stuck on a hard paragraph, staring out the window to think. The signal drops, the band buzzes, and it interrupts the exact moment of focus it was meant to protect. Baselines also differ from person to person, so one fixed line would nag some users constantly and never notice others.
So a dip only becomes a candidate, measured against the user's own baseline from the first minutes of a session. It's not a trigger yet.
Q2
A fixed delay sounds like the answer, but the focus group pointed somewhere better. Many people already notice and recover on their own. A cue that lands during that recovery takes the credit and adds noise.
So each candidate opens a short window, around 20 seconds in the prototype. If the reading comes back before the window closes, the timer quietly resets. Only a change that holds counts as drift.
Q3
My first version was a nudge that simply said "focus."
But a bare reminder interrupts without helping. Picture someone who has been rereading the same page for a minute. The band buzzes, they realize they drifted, and now they also have to decide what to do about it. The cue adds a decision on top of the distraction.
So every cue comes with one small next step. When sustained low readings trigger a reminder, the product suggests a brief reset, like a short breathing round or a quick stretch, then brings the user back to their task. The interruption gets a clear purpose: notice, reset, resume.
Putting it together
Each decision removes a reason to interrupt, until the only cue left is one worth acting on.
Bringing focus support into everyday life.
The feedback logic defines when to step in and how to guide people back. Next, the question became how a wearable and an app could carry that experience together.
A working prototype, built to test the idea.
To test the experience with real people, I built a headband that could be worn through a full work session. I treated it as a design tool, not a product to ship: every hardware choice was judged by what it meant for the person wearing it.
How the signal travels
Designing for the wearer
+Technical details
Board headers: P1 EEG electrode, P3 UART serial, P4 power. The Arduino reads packets at 57600 baud, checks the sync bytes and checksum, then extracts signal strength, attention and meditation values. A Velcro band adjusts to different head sizes. View the Arduino code ↗
An app that knows when to step back.
The band handles the moment focus slips. The app supports everything around it: getting started, choosing a reset, and looking back afterwards. Every screen traces back to something the research or the key decisions asked for.
User flow
One session is a short loop. The only interruption, the cue, waits behind the 20-second rule from Q2. Everything else waits until focus is over.
Setting up
A short calibration sets your own baseline. Then two choices, and you're focusing.
During focus
Readable at a glance, so the phone can stay face down. Colour and rhythm shift with each state, and the band only steps in after 20 s.
When focus slips
Each cue suggests the reset that fits what's behind the drop. One tap to start, skip anytime.
After the session
Details wait until focus is over.
Visual language
Warm paper tones and one calm blue, so the screen never competes with the task. Violet, orange and pink only appear when the state changes: drifting, running low, or tense. Each one matches its own reset.
See it in motion
One session from start to finish: plan, focus, a gentle nudge when attention drifts, a short reset, and the look back afterwards.
Did it help?
I ran a within-subject pilot: 6 participants each completed the same learning task with and without the device, then rated their focus. Numbers below are group averages.
| Measure | Without device | With device |
|---|---|---|
| Task time | 35 min | 29 min |
| Accuracy | 88% | 90% |
| Self-rated focus | 6 / 10 | 9 / 10 |

