Effects of e-map format and sub-window on navigation performance and glance behavior Hsin-Chieh Wu a *, Ching-Torng Lin b , Ting-Yen Chien b a Department of Industrial Engineering and Management, Chaoyang University of Technology, Taiwan b Department of Information Management, Da-Yeh University, Taiwan * Corresponding author. E-Mail address: hcwul@cyut.edu.tw (H.-C. Wu) With the advances in global position system (GPS), automobile manufacturers have begun to produce intelligent transportation systems to assist drivers, such as in-vehicle navigation systems. These navigation systems combine an electronic map (e-map) and permanent roadway signs to inform drivers of their current locations and traffic information visually or aurally. To date, the moving e-map is still the primary medium for navigation or GPS applications; it can’t be completely replaced by voice guidance. An on-road driving experiment was conducted to investigate the effects of e-map format and sub-window on navigation performance and glance behavior while using an in-vehicle navigation system. Twenty participants navigated an urban route using the navigation display under 2D or 3D e-map conditions, and either with a sub-window or not. Driver navigation errors and visual glance data were collected during the trials. Results show that a sub-window reduced navigation errors but increased the time spent glancing to the visual display. There was no difference in performance between the 2D and 3D e-map conditions. However, 3D e-map led to more visual glance behavior to the navigation display than 2D one. The wider implications for the design of navigation visual displays are discussed. Keywords: Driving performance; Interface design; Navigation display INTRODUCTION With the advances in global position system (GPS), automobile manufacturers have begun to produce intelligent transportation systems to assist drivers, such as in-vehicle navigation systems. These navigation systems combine an electronic map (e-map) and permanent roadway signs to inform drivers of their current locations and traffic information visually or aurally. They offer a technological solution to driver navigation in an unfamiliar area. While drivers can benefit from receiving route guidance information and traffic status, they have to face heavier visual loading at the same time. Besides, when receiving related information provided by the navigation system, drivers need to keep their eyes off the road, which may seriously influence driving safety. Distraction from the primary driving task is one of main causes of traffic dangers (Wierwille, 1995). Previous studies have also reported that the increase of visual demand for using a navigation system could bring negative impacts on driving safety and performance (Ross and Burnett, 2001; Toshiaki, et al., 2003; Williams and Helbig, 2006). Therefore, one major requirement for display-based navigation systems is that information must be readable and understandable fast and reliably (Baumann, et al., 2004). The e-map presentation technique has been developed from two-dimensional (2D) to three-dimensional (3D) formats (Van Orden and Broyles, 2000). As we know, the actual world is spread in the 3D space and thus the 3D format is more like ‘real world’ representation that may imply better human perception and performance. However, previous studies have revealed that 3D displays are not markedly superior to 2D ones for the air combat or air traffic control tasks (Tham and Wickens, 1993; Van Orden and Broyles, 2000). These studies mostly focused on the performance of pilots or air traffic controllers; little research has aimed at evaluating the differences in driving performance between 2D and 3D formats. There is a need to evaluate the effect of e-map format on driving performance and visual demand while using a navigation display. In recent years, the multi-windows interface technology has made the navigation display capable of simultaneously presenting both a main-window and a sub-window. The sub-window is usually designed to highlight the information of turn direction (i.e. present the region of intersection at large scale next to the main window) when approaching an intersection. Combining a main-window with a sub-window, the navigation display could provide noticeable and redundant route guidance that might result in better navigational performance than a single-window display. But an extra sub-window may impose more visual demand on the users. The effect of sub-window on driving performance is still not fully understood because most navigation-related studies used single window display only (Liu, 2000; Streeter, et al., 1985; Sikanen, et al., 2005; Van Orden and Broyles, 2000). Besides, how sub-window interacts with drivers to affect attention distraction remains to be evaluated.