Social Dilemmas in Vehicle Automation.

The fundamental building blocks for automating vehicles have been in development for many years, making vehicle automation a near-term reality (Barth et al., 2014). The interest and the expectations, with which the scientific world, the markets and urban societies anticipate a full scale launch of ‘driverless’ or ‘semi-automated’ vehicle technology are higher than ever before. This is because automated vehicles have, in theory at least, the potential to completely transform urban development as we know it, with a revolution in ground transport, regulations permitting, that could dramatically change the landscape of cities around the world and have an enormous economic, social, spatial, and mobility impact (Alessandrini et al., 2015). Although recent studies showed that a priori acceptability of fully automated cars could be likely for many drivers today (Payre et al., 2014) things could get far more complex. This is because, despite some initial encouraging technical results, the implementation of vehicle automation, will not be straightforward, predictable or unproblematic; there is a wide spectrum of social dilemmas and complicated human factors issues that may arise for such an ‘untested’ and ‘powerful’ innovation. One of the likeliest consequences of vehicle automation on the average ‘driver’, for instance, could possibly be the loss of situation awareness and loss of the skills needed to perform the automated functions manually (Parasuraman et al., 2000). Previous research has revealed problems with driving performance in situations with automation failures and attributed this to drivers being out-of-the-loop (Strand et al, 2014). Another issue with the introduction of on-board technologies resides in the fact that drivers can react in unexpected ways to the introduction of new systems, a phenomenon defined as ‘behavioural adaptation’ (Gouy et al., 2014). The formation of a mixed traffic situation where vehicles equipped with automated systems taking over the lateral and longitudinal control of the vehicle will interact with unequipped vehicles that are not fitted with such automated systems is another likely future scenario that could arise societal dilemmas (Gouy et al., 2014). Also, future automated vehicles will encounter situations where the ‘right’ action is morally or legally ambiguous; allocating risks without a human driver’s oversight is an issue of ethical dimension, particularly in instances where an automated vehicle would not be able to avoid crashing (Goodal, 2014). The paper aims to examine some of these social dilemmas via a state-of-the-art review of the existing literature but more importantly, perhaps, via the analysis of a series of semi-structured in-depth interviews. The interviews will be testing future hypothetical scenarios to an audience spanning from researchers working in the field of automation to people that have never discussed about ‘driverless’ vehicles before in any formal sense. This approach is expected to deliver an intriguing mix of results that could have a more generalizable nature when trying to frame, to the degree that is possible, some of the inescapable, but yet somewhat unknown, realities in vehicle automation