Abstract
Vehicle-based warnings that speed road hazard detection can reduce collision incidence and severity. However, no technology is perfect, and it is critical to understand the impact of incorrect cues on hazard detection. This study, conducted between September 2022 and September 2023, examined the impact of non-spatial hazard warning auditory cues on licensed drivers' ability to localize hazards (situations requiring immediate response to avoid a collision) in real road footage when cues are mistimed (Experiment 1) and when cues include false alarms (Experiment 2). In Experiment 1, we varied the duration between cue and hazard onset and found that earlier cues speeded responses more than later cues, and warning cues reduced response time regardless of timing. However, each trial included a hazard, whereas hazards are rare on the road. In Experiment 2, we added false alarm warnings and hazard-absent trials in two cue reliability conditions (80% and 50%), and these cues did not significantly affect hazard localization performance regardless of reliability. Although earlier auditory temporal warnings can speed hazard localization, these benefits disappear in the presence of false alarms in attentive drivers and suggest that classic cueing results may not necessarily translate to dynamic natural scenes with ambiguous targets onsets.