"I stepped into a puddle": Non-Visual Texting in Nomadic Contexts
Pegah Karimi, Erin Brady, Aqueasha Martin-Hammond, Davide Bolchini · 2023 · Proceedings of the 20th International Web for All Conference (W4A '23) · doi:10.1145/3587281.3587285
Summary
This study investigates the under-explored practice of texting by people who are blind and visually impaired (BVI) while navigating outdoors — what the authors term "nomadic contexts." While previous research has examined accessible text entry methods (screen readers, speech input, screenless typing) and BVI navigation separately, the interplay between texting and outdoor navigation has received little attention. The researchers conducted semi-structured interviews with 20 BVI participants (9 totally blind, 10 legally blind, 1 with light perception; ages 27-69; 12 female, 8 male) recruited from Midwest-based non-profit organizations. Participants used diverse assistive technologies including Siri, Be My Eyes, Seeing AI, Aira, and various wayfinding apps (Google Maps, Apple Maps, BlindSquare, Soundscape). The study identified four recurrent situations where BVI users text outside the home: walking to a destination, waiting for public transportation, riding in a vehicle, and approaching a point of interest. In each situation, texting practices differ significantly based on available hands, safety concerns, privacy needs, and environmental factors. For example, when walking, most users (n=12) stop and stand still to text; when riding in vehicles with strangers, they avoid voice-based tools for privacy; when waiting at bus stops, they often fold their cane to free their hands.
Key findings
The study identified nine types of situational information that BVI users need while texting outdoors: location, landmarks, obstacles, bus schedule, ETA, distance to destination, weather, unfamiliar sounds, and traffic jams. Location and landmarks were critical across all four situations, while obstacles were important in all situations except in-vehicle. The researchers developed three personas representing distinct texting behavior patterns: "Texting Anywhere" (confident multi-taskers who text while walking with Siri), "Texting for Updates" (users who frequently share location updates, concerned about getting lost), and "Texting if Necessary" (safety-conscious users who text only when needed, preferring to find a seat first). Key challenges included application switching (juggling between navigation apps, assistive technology apps like Seeing AI, and messaging apps), difficulties operating texting apps with VoiceOver while managing a cane or guide dog, and reduced situational awareness. One participant's account of stepping into a puddle while dictating a text dramatically illustrates the safety risks. Participants proposed three modes of situational feedback to integrate with messaging apps: text-based sequential cues, sound effects mapped to environmental hazards, and tactile/vibration alerts for safety-critical information.
Relevance
This research fills an important gap by examining how BVI users manage the competing demands of navigation and communication simultaneously — a dual-task challenge that sighted users also face but which is fundamentally different without vision. The findings have direct implications for mobile messaging app developers: integrating situational awareness features (obstacle alerts, location sharing, transit updates) directly into texting interfaces would eliminate the dangerous and cumbersome app-switching that BVI users currently endure. The three feedback modalities proposed (text, sound, tactile) offer a concrete design framework, with participants noting that different modalities suit different situations — sound effects for quick hazard recognition, vibration for safety alerts in noisy environments, and text for detailed information. The persona-based analysis is useful for designers building inclusive messaging features, though the study is limited by its interview-only methodology without field observation. Future work combining observation with interviews would provide richer insights into actual multitasking behaviors.
Tags: blind and low vision · text entry · mobile accessibility · navigation · assistive technology · touchscreen accessibility · VoiceOver · speech recognition · situational awareness · public transportation · independent living