Comparing trust in a Human Driver Versus an Automated Vehicle Following a Dangerous Situation

With automated vehicles becoming increasingly widespread, the modern transportation landscape is likely to change. Compared to manually operated vehicles, automated vehicles are expected to offer a range of benefits, from reduced environmental impact to increased road safety. However, the success of this technology depends on whether human passengers can learn to interact with it appropriately. The process of trust calibration is important for this, as appropriately calibrated trust enables a passenger to rely on the vehicle when it is warranted but intervene in situations where the capabilities of automation fall short. Therefore, investigating factors affecting trust is crucial for developing systems that facilitate successful human–automated vehicle partnerships. To learn more about the dynamics of trust development during a drive in an automated vehicle, we conducted this experimental study. In it, two groups of participants underwent a driving simulation in a virtual reality...Show moreWith automated vehicles becoming increasingly widespread, the modern transportation landscape is likely to change. Compared to manually operated vehicles, automated vehicles are expected to offer a range of benefits, from reduced environmental impact to increased road safety. However, the success of this technology depends on whether human passengers can learn to interact with it appropriately. The process of trust calibration is important for this, as appropriately calibrated trust enables a passenger to rely on the vehicle when it is warranted but intervene in situations where the capabilities of automation fall short. Therefore, investigating factors affecting trust is crucial for developing systems that facilitate successful human–automated vehicle partnerships. To learn more about the dynamics of trust development during a drive in an automated vehicle, we conducted this experimental study. In it, two groups of participants underwent a driving simulation in a virtual reality environment, one being driven by an automated vehicle and the other by a human driver. Both groups experienced a dangerous traffic rule violation during the drive. We assessed the trust levels during the drive and found that trust significantly dropped after the rule violation with no full recovery afterwards. We observed a similar pattern in both groups, with no significant differences between the human and the automated vehicle conditions both in terms of mean trust scores and pattern of trust development. Furthermore, we assessed the relationship between trust and extroversion, but no significant correlation was found. Our findings suggest that trust calibration is a dynamic process, with participants continuously updating their expectations regarding the vehicle’s safety and capabilities. This knowledge contributes to the understanding of factors influencing human–automation relationships and can help aid the integration of automated vehicles in today’s transportation landscape.Show less

Leiden University Student Repository

Documents

In Collections

Comparing trust in a Human Driver Versus an Automated Vehicle Following a Dangerous Situation