LEDs on a strip, when turned on and off in a specific order, result in the perception of apparent motion (i.e. beta movement). In the automotive domain such chase lights have been used to alter drivers' perception of driving speed by manipulating the pixel speed of LEDs. We argue that the perceived velocity of beta movement in the peripheral view is not only based on the actual pixel speed but can be influenced by other factors such as frequency, width and brightness of lit LED segments. We conducted a velocity matching experiment (N=25) by systematically varying these three properties, in order to determine their influence on a participant's perceived velocity in a vehicle mock-up. Results show that a higher frequency and stronger brightness increased perceived velocity, whereas segment width had no influence. We discuss how findings may be applied when designing systems that use beta movement to influence the perception of ambient light velocity.

LED visualizations in the car are primarily used to direct the driver's attention to obstacles or hazards in the environment. An ambient LED arrangement can indicate the horizontal position of outside objects without obstructing the driver's field of view. How accurately the LEDs represent the spatial position of an outside object and how well it can be distinguished from other nearby objects is influenced by several factors. Our aim is to increase the spatial accuracy of a windshield-mounted horizontal LED visualization. Our approach is to use a calibration procedure to increase this accuracy. In an experimental study in our driving simulator, we found that calibration increased the spatial accuracy and, as a result, the recognition reliability (in our setup from 67% to 95%) and that a 3-point calibration is sufficient in comparison to more calibration points.

In order to support drivers to maintain a predefined driving speed, we introduce ChaseLight, an in-car system that uses a programmable LED stripe mounted along the A-pillar of a car. The chase light (i.e., stripes of adjacent LEDs that are turned on and off frequently to give the illusion of lights moving along the stripe) provides ambient feedback to the driver about speed. We present a simulator based user study that uses three different types of feedback: (1) chase light with constant speed, (2) with proportional speed (i.e., chase light speed correlates with vehicle speed), and (3) with adaptive speed (i.e., chase light speed adapts to a target speed of the vehicle). Our results show that the adaptive condition is suited best to help a driver to control driving speed. The proportional speed condition resulted in a significantly slower mean speed than the baseline condition (no chase light).

In demanding driving situations, the front-seat passenger can become a supporter of the driver by, e.g., monitoring the scene or providing hints about upcoming hazards or turning points. A fast and efficient communication of such spatial information can help the driver to react properly, with more foresight. As shown in previous research, this spatial referencing can be facilitated by providing the driver a visualization of the front-seat passenger's gaze. In this paper, we focus on the question how the gaze should be visualized for the driver, taking into account the feasibility of implementation in a real car. We present the results from a driving simulator study, where we compared an LED visualization (glowing LEDs on an LED stripe mounted at the bottom of the windshield, indicating the horizontal position of the gaze) with a visualization of the gaze as a dot in the simulated environment. Our results show that LED visualization comes with benefits with regard to driver distraction but also bears disadvantages with regard to accuracy and control for the front-seat passenger.

This paper presents a design exploration of full-body interaction games played in cars. It describes how we have designed, implemented, and evaluated the core experiences of three different games, which were all aimed at making sitting properly more fun for players/children while travelling by car. By making the restricted body an integral part of gameplay, we hope to, as a side product of gameplay, bring about the best and also most safe body posture for young players/children travelling by car, i.e., sitting reasonably upright and still in their child seat with their head leaning back on the neck rest. Another outcome of this could also be an overall safer situation in the car, in that children not sitting still in their child seats while being driven might be stressful for the driver. By presenting the details of our design efforts in this particular design context, we hope to add also to the knowledge we, in HCI, have for how to design bodily experiences with technology at large.

User Experience (UX) studies in the automotive context are essential for a good interaction design in the car. But there is a lack of appropriate methods to investigate the user's experience in-situ in a natural setting. In this paper we introduce computerized experience sampling in the car. We present an ESM tool based on the OBD-II interface of a car in combination with an android smartphone and describe issues and challenges we experienced while applying our tool in an initial UX study.

Automobiles are environments rich of data that have a high potential to be used as input for research and design. Various difficulties accessing car data exclude HCI experts from making use of this data in novel interfaces and research projects. We present the CarDaT (Car Data Toolkit) that uses Android smartphones to provide multidimensional sensor data in a minimal invasive way. CarDaT combines smartphone sensor data with data sources like OBD-II as well as other easily available remote data (e.g., weather). This data and the provided connectivity enable researchers to gather data on human behavior and designers to create novel context-aware interface solutions. Thus, CarDaT offers a low-cost, manufacturer independent and scalable in-car agile prototyping and research environment. In this paper we describe how we used smartphones in the CarDaT as tools for automotive research and design. We demonstrate potentials of CarDaT by describing three applications that we developed with the toolkit, namely rear seat games, an experience sampling study, as well as an experiment using car data in a driver distraction study to inform design.