For the first time in history, automated vehicles (AVs) are being deployed in populated environments. This unprecedented transformation of our everyday lives demands a significant undertaking: endowing complex autonomous systems with ethically acceptable behavior. We outline how one prominent, ethically relevant component of AVs—driving behavior—is inextricably linked to stakeholders in the technical, regulatory, and social spheres of the field. Whereas humans are presumed (rightly or wrongly) to have the “common sense” to behave ethically in new driving situations beyond a standard driving test, AVs do not (and probably should not) enjoy this presumption. We examine, at a high level, how to test the common sense of an AV. We start by reviewing discussions of “driverless dilemmas,” adaptions of the traditional “trolley dilemmas” of philosophy that have sparked discussion on AV ethics but have limited use to the technical and legal spheres. Then, we explain how to substantially change the premises and features of these dilemmas (while preserving their behavioral diagnostic spirit) in order to lay the foundations for a more practical and relevant framework that tests driving common sense as an integral part of road rules testing.

In a period of fewer than 10 years, the quest for self-driving vehicles, also referred to as autonomous vehicles (AVs) or driverless cars, has become one of the biggest technology races in the world, with tens of billions of dollars poured into companies and start-ups. The goal is an on-road, consumer-driverless car: whether owned by individuals or part of a centralized ride-sharing fleet, this is the area where the majority of investment has occurred. However, AVs have been around for much longer in other fields, such as mining, which share some but not all of the same technical challenges faced by on-road AVs. In this article, we provide an overview of the key technical challenges and solutions for both onand off-road AVs, with a focus on one of the key unsolved challenges-interaction with vulnerable road users (VRUs).

For the first time in history, autonomous vehicles (AVs) are being deployed in populated environments. This unprecedent transformation of our everyday lives demands and relies upon a substantial requirement: endowing complex autonomous systems with ethically acceptable behavior. We outline how ethics is inextricably linked to several stakeholders in the technical, regulatory, and social perception spheres of autonomous vehicles. At bottom, the main difference between humans and AVs is that humans are presumed to have the ‘ethical common sense’to deal with new driving situations beyond a standard driving test, whereas for AVs we need to test this ethical sense too. We flesh out, at a high level, what such a test could look like. We start by reviewing the proposal from studies of ‘driverless dilemmas’, an adaption of the traditional ‘trolley dilemmas’ of philosophy that have sparked helpful discussion on AV ethics, yet with limited use to the technical and regulatory spheres. Then, we explain how to substantially change the premises and features of these dilemmas (while preserving their behavioral diagnostic spirit) in order to lay the foundations of an ethics test for self-driving cars.

The alarm has been raised on so-called driverless dilemmas, in which autonomous vehicles will need to make high-stakes ethical decisions on the road. We argue that these arguments are too contrived to be of practical use, are an inappropriate method for making decisions on issues of safety, and should not be used to inform engineering or policy.

Orienting a logo upward or downward may seem like an arbitrary graphic design decision, but we propose that it can have important implications for consumer judgments. In particular, we find across four experimental studies and a content analysis that diagonal direction can convey different levels of activity with upward—or ascending—diagonals conveying greater activity and effort than downward—or descending—diagonals. Consequently, when the context highlights the benefits of activity (vs. passivity), upward (vs. downward) diagonals lead to more favorable product judgments, greater product efficacy beliefs, and greater post-consumption satisfaction. Furthermore, we provide process evidence that perceived product efficacy beliefs mediate these effects, and that the effect is strongest when the object being visually oriented is text rather than images. These findings are particularly important in light of our content analysis findings that diagonal orientation is a relatively underutilized design feature. Collectively, our findings suggest that firms should use upward diagonals when the product context highlights a favorable view of activity. Otherwise, the firm should use downward diagonals, especially when the product context encourages consumers to view passivity favorably.

The most influential models of face perception over the past three decades have regarded the generation of an at least partially invariant representation as the necessary first step in most face processing tasks. Correspondingly, models of “high-level” face perception tasks such as personality judgment have relied on features likely to be measured or preserved in an invariant representation (e.g. physiognomy, pose). Several recent papers have proposed deep convolutional networks as a biologically-plausible mechanism for creating invariant representations, and have shown that those networks, when trained on recognition tasks, are able to predict activity in anterior portions of the ventral visual stream. We have created a series of models that predict human judgments of personality traits (such as trustworthiness and dominance). In these models, performance of “shallow” v1-like feature sets compares highly favorably to that of deep convolutional feature sets. By probing the feature space of these shallow models we show that the best-performing models’ performance is driven by low-level features that are highly variable, even among images of a single individual. These features can be modified to change the relevant attribute rating for a given image without changing humans’ subjective perception of invariant face characteristics such as identity or physiognomy. These results call into doubt the necessity or pre-eminence of an invariant face representation in the judgment of personality traits from face images. This in turn suggests that models of the face perception system – and of personality trait judgments – that operate on invariant representations of the type efficiently generated by biologically plausible deep convolutional networks may not capture the types of features most relevant for some high-level face perception tasks. Meeting abstract presented at VSS 2015