Effects of Visual Field Loss on Drivers' Explicit and Implicit Initial Trust Preferences: Full Driving Automation versus Themselves

Level 5 automated driving systems (L5 ADS) require no human operation and thus offer potential mobility benefits for individuals with visual field loss (VFL). This study investigates how VFL influences individuals' explicit and implicit initial trust preferences between themselves and L5 ADS, defined as the relative tendency to trust L5 ADS over one's own driving capability, before real-world experience with such systems. Two studies were conducted. Study 1 employed a laboratory experiment in which VFL was simulated in healthy drivers to examine a causal relationship between the presence of VFL and initial trust preferences. To address the limitations of standardized VFL simulation, given that individuals with real VFL exhibit variability in disease progression and long-term adaptive strategies, Study 2 compared drivers with real VFL to healthy drivers. Results showed that, relative to the no-VFL condition, simulated VFL increased drivers' initial trust in L5 ADS and reduced their driving self-efficacy, leading to a stronger explicit initial trust preference for L5 ADS over themselves. In contrast, implicit initial trust preferences remained unchanged. In Study 2, drivers with real VFL demonstrated an implicit initial trust preference for themselves, which differed significantly from that of healthy drivers, who showed an implicit initial trust preference for L5 ADS. However, no significant group differences were observed in initial trust in L5 ADS, driving self-efficacy, or explicit initial trust preferences. Both groups explicitly preferred themselves. These findings suggest that assessments of initial trust attitudes toward L5 ADS among VFL populations should combine explicit and implicit measures to avoid misjudgment in contexts such as policymaking, market forecasting, and user evaluation and education. • Implicit trust preferences toward Level 5 ADS often deviate from explicit ones. • Both visual field loss drivers and healthy drivers explicitly trusted themselves. • Visual field loss drivers also implicitly trusted themselves. • Healthy drivers conversely implicitly trusted Level 5 ADS more than themselves. • Simulated visual field loss only increased explicit trust preference, not implicit.