Collaboration and Assistive Technology: Facilitating Joint Awareness for Noise Sensitivity
Emani Hicks, Luc Rieffel, Ariya Gowda, Aehong Min, Gillian R Hayes · 2026 · Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI '26) · doi:10.1145/3772318.3793203
Summary
People with noise sensitivity (PWNS) - including those with misophonia, hyperacusis, and tinnitus, and overlapping strongly with the autistic community (estimated 45% prevalence) - experience everyday sounds as painful, distressing, or overwhelming. Existing supports are patchy: earplugs and headphones are not always socially acceptable, desensitisation therapies are slow, and prior HCI work has focused on self-monitoring tools that leave companions (parents, partners) out of the loop. This study frames noise sensitivity management as a collaborative care problem and asks whether mobile and wearable technology can support joint awareness - the shared understanding between a PWNS and their allies about what triggers dysregulation, how the body responds, and what helps. The authors built AudioBuddy, an iOS + Apple Watch app with four integrated features: mood check-ins across four categories and sixteen emotions, personalised heart-rate and decibel-level thresholds with real-time alerts, a 'Notify Ally' function that can email or iMessage a companion with location and vital signs, and a coping-activity toolbox integrated with Apple Calm. They deployed AudioBuddy as a technology probe for two weeks with eleven PWNS (ages 11-62) and seven companions - mostly parent-child dyads and two romantic partnerships - collected via the Sussex Misophonia Scale and Hyperacusis Assessment Questionnaire, and conducted four rounds of interviews analysed via reflexive thematic analysis with affinity diagramming.
Key findings
Three themes structured the findings. (1) Data-driven reflection fostered awareness: participants identified previously invisible patterns, e.g. N1 realising her heart rate spiked in a store that 'wasn't as noisy' as she thought, and C1 visualising Disneyland as a cumulative noise load. Critically, companions reported growing empathy ('I would say I have a little bit more grace for her' - N1's mother), suggesting data-driven reflection reshapes relational dynamics, not just self-knowledge. (2) Noise alerts validated lived experience: eight of eleven PWNS had HAQ scores in the loudness-problem range, and AudioBuddy's alerts corroborated subjective reports ('It helped to prove his case that his music was loud' - C3). But the Apple Watch only measures decibels, so participants consistently flagged that problematic sounds are often about pitch, rhythm, timbre, or specific sources (chewing, dog panting, mouth noises) rather than volume - a fundamental limit of consumer wearables. (3) Mood logging and Notify Ally supported joint awareness unevenly: check-ins helped both PWNS and companions pause and reflect, and companions often wanted to *passively view* the PWNS's data. But dysregulated moments were exactly when PWNS could not navigate the app to send an alert, and the alert wording felt 'dramatic' in contexts that were uncomfortable but not urgent. Participants surfaced tensions between self-advocacy burden, privacy, hypervigilance risk, and surveillance - for example C4's worry that visibility could feel like 'spying' on his child.
Relevance
For accessibility practitioners this paper is a clear case for moving beyond individualistic self-tracking toward what the authors call relational and structural-level intervention. Four concrete design implications are directly actionable: (DI1) consent-based data sharing with configurable granularity rather than all-or-nothing access; (DI2) time-delayed summarised reports to reduce surveillance feel while preserving joint awareness; (DI3) situated interfaces in shared spaces (classrooms, workplaces, NICUs) that communicate general noise conditions without stigmatising any one individual; (DI4) sound recognition that captures characteristics beyond loudness - pitch, timbre, specific triggers. The tension the paper surfaces is widely applicable across assistive tech: systems designed to enable self-advocacy can inadvertently burden the very people they aim to help, particularly children in school settings with device restrictions, and dysregulated adults whose executive function is compromised. Limitations are significant - two weeks is short, participants self-selected as motivated, devices were loaned to Android users who had to switch platforms, and the Notify Ally feature saw very low uptake. The paper is most useful as a design-lessons artefact, not a validation of AudioBuddy itself.
Tags: noise sensitivity · misophonia · hyperacusis · sensory processing · autism · neurodivergence · wearable technology · mobile accessibility · assistive technology · collaborative care · self-regulation · technology probe · qualitative research