MaskSound: Exploring Sound Masking Approaches to Support People with Autism in Managing Noise Sensitivity
Anna Y Park, Andy Jin, Jeremy Zhengqi Huang, Jesse Carr, Dhruv Jain · 2024 · Proceedings of the 26th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS 2024) · doi:10.1145/3663548.3675656
Summary
This paper investigates sound masking as an alternative strategy for helping autistic individuals manage noise sensitivity in everyday environments. Unlike noise cancellation or isolation approaches that block out environmental sounds entirely, sound masking introduces ambient background sounds to reduce the perceptual salience of disruptive noises while preserving situational awareness. The researchers conducted two studies with autistic participants. Study 1 involved 10 autistic adults who evaluated eight different sound masks—white noise, brown noise, pink noise, calming water, rain, wind, birdsong, and coffee shop ambiance—across five common noise contexts: traffic, construction, speech babble, eating sounds, and baby crying. Participants rated each mask on effectiveness, annoyance, and preference while listening through headphones. Study 2 used a Wizard-of-Oz prototype with 9 autistic participants who experienced real-time, context-adaptive sound masking in realistic multi-room scenarios featuring overlapping noise sources. The research was motivated by the limitations of existing approaches: noise-cancelling headphones reduce awareness of surroundings, earplugs can be uncomfortable, and complete sound isolation limits social engagement. Sound masking, borrowed from architectural acoustics and tinnitus therapy, offers a middle ground that softens the impact of triggering sounds without eliminating environmental audio entirely. The participatory approach centered autistic voices throughout the design process, recognizing that noise sensitivity is a significant barrier to daily activities, social participation, and mental wellbeing for many autistic people.
Key findings
Natural sounds, particularly calming water and rain, were consistently preferred over synthetic noise colors like white, pink, and brown noise across most contexts. However, preferences were highly individual and context-dependent—no single mask worked universally well. Some triggering sounds proved harder to mask than others: eating sounds and baby crying were particularly challenging because participants' distress was driven by emotional and sensory associations rather than purely acoustic properties. Participants strongly desired control over mask selection, volume, and activation timing, rejecting fully automated systems in favour of user-initiated masking with optional context suggestions. The concept of layering multiple masks simultaneously emerged as a popular request, with participants wanting to combine sounds like rain with birdsong for greater effectiveness. A "sound profile" concept also emerged, where users envisioned saving preferred mask configurations for recurring contexts like commuting or dining out. In Study 2, participants found that sound masking reduced anxiety and sensory overload while maintaining enough environmental awareness to navigate social situations and detect important sounds like alarms. Gradual volume transitions were preferred over abrupt changes, which could themselves be startling.
Relevance
This research opens a promising new direction for sensory accessibility technology that goes beyond the binary of hearing everything or blocking everything out. For accessibility practitioners, the findings highlight the importance of customisation and user agency in assistive tools for autistic users—one-size-fits-all solutions are insufficient given the highly individual nature of sensory sensitivities. The sound masking approach could inform the design of mobile apps, smart headphones, and environmental sound systems in workplaces, schools, and public spaces. The study also demonstrates the value of Wizard-of-Oz prototyping for exploring complex assistive technology concepts before committing to full implementation. A key limitation is the small sample size and controlled lab setting; real-world deployment would face challenges around ambient sound detection, latency, and battery life. The emotional dimension of noise sensitivity—where certain sounds trigger distress regardless of volume—suggests that acoustic masking alone may need to be complemented by other strategies.
Tags: autism · noise sensitivity · sound masking · sensory accessibility · auditory environment · assistive technology · participatory design