Addressing the Situational Impairments Encountered by Firefighters through the Design of Alerts
Flynn Wolf, Priyanka Soni, Ravi Kuber, Dianne Pawluk, Brian Turnage · 2019 · Proceedings of the 16th International Web for All Conference (W4A) · doi:10.1145/3315002.3317556
Summary
This paper describes a structured multi-step approach to designing multimodal alerts for firefighters who experience severe situationally-induced impairments and disabilities (SIIDs) at fire scenes — thick smoke restricting vision, loud noise from combustion, sirens, and power tools blocking auditory communication, and heavy protective gear limiting tactile sensitivity. The research followed four steps: (1) contextual interviews with six safety officers and senior fire personnel from two districts to understand current alerting practices and gaps; (2) participatory design workshops with six groups of 4-8 firefighters and interface designers at fire stations, using illustrated scenario-based prototyping; (3) lab-based evaluation of the designed alerts under simulated conditions; and (4) planned evaluation at a fire training academy under realistic conditions. The contextual interviews revealed three critical information gaps not addressed by existing alerts: environmental temperature (which can rapidly reach flashover levels), gases/toxins/carcinogens present, and core body temperature/cardiac stress of the firefighter. From participatory design, three alert types were developed across three modalities: visual (LED strips inside the facemask in peripheral vision, using color coding — blue for cardiac, magenta for external temperature, cyan for core temperature), auditory (bone conductance speaker on the cheek, using distinct pitch patterns — chirpUp, tripleTone, and rough sequences), and tactile (vibration motor on the forehead brow area, using different vibration patterns). Each alert had three severity levels (OK, caution, critical) encoded through flash rate/interval patterns, with increasing repetition frequency for higher urgency.
Key findings
Lab evaluation with 12 participants wearing facemasks with simulated smoke occlusion and fire-scene audio (82-109 dBA) showed that visual alerts were most accurately recognized for threat level (99.4% accuracy) compared to auditory (92.9%) and tactile (85.7%). For threat type identification, visual was again most accurate (100%) versus auditory (98.2%) and tactile (73.2%). Visual alerts were also fastest (4.56 seconds) compared to auditory (6.6s) and tactile (10.58s). However, tactile cues imposed the highest perceived cognitive workload (1.95/5) compared to auditory (1.38) and visual (1.35). The poorer tactile performance was attributed to limited user familiarity (firefighters have extensive training with visual and auditory alerts but not tactile), actuator limitations of pancake motors, and potential interference from physical exertion. Despite the lower quantitative performance, subjective feedback was positive about tactile alerts — firefighters valued the concept of a "private channel" that doesn't distract nearby teammates and works when vision and hearing are both blocked. Errors most commonly occurred distinguishing caution from critical levels in auditory and tactile modalities. A key participatory design finding was that firefighters demanded extreme simplicity — they cautioned against too many alert mappings, rejected complex encoding schemes, and insisted that alerts be immediately intuitive under cognitive and physical duress. They also raised practical constraints: metal and plastic items near the body can heat up and burn skin, heavy turnout gear attenuates signals, and any additions that slow the donning process would be rejected.
Relevance
This research extends the concept of accessibility beyond the traditional scope of people with permanent disabilities to encompass anyone whose sensory or motor capabilities are temporarily restricted by their environment. Firefighters represent an extreme case of SIIDs where multiple channels (visual, auditory, motor) are simultaneously impaired, creating conditions analogous to deafblindness. The findings have broader implications for accessible alert design in any context where users may face sensory restrictions — construction sites, industrial environments, underwater operations, or even everyday situations like driving or using devices in bright sunlight. For accessibility practitioners, the participatory design methodology is particularly instructive: involving the target user community as co-designers (rather than just evaluators) produced practical insights that researchers would not have anticipated — like the importance of not adding to equipment donning time, the risk of thermal burns from wearable electronics, and the need for alerts that are intuitively meaningful under extreme cognitive load. The finding that visual alerts (via peripheral LED strips) outperformed both auditory and tactile alternatives under simulated fire conditions challenges assumptions about which modalities work best when vision is impaired, suggesting that even partially restricted vision can be highly effective for simple color-coded peripheral cues.
Tags: situational impairment · situational disability · haptic feedback · multimodal interaction · participatory design · firefighting · alert design · situational awareness · wearable technology · tactile feedback