Search

Elevation change is critical for choosing a route when bicycling. We detail an empirical study in which we comparatively evaluated three linear symbolization types (color hue, color-coded arrows, and elevation profiles) to depict elevation change in bicycle maps for two common bicycle route planning tasks: relative height detection and slope identification. Participants performed most accurately with the color-coded arrows for relative height detection tasks, whereas symbolization did not significantly influence map-use performance for slope identification. Participants preferred the elevation profile, in spite of their lowest performance with this method overall. Our rare empirical findings offer much needed new insights into the function and appropriateness of common elevation symbolization methods, specifically to identify elevation change in bicycle route planning tasks in urban areas where map display real estate is already very limited.

GIScience 2016 Short Paper Proceedings Walk and Learn: An Empirical Framework for Assessing Spatial Knowledge Acquisition during Mobile Map Use A. Brugger 1 , K.-F. Richter 1 , S. I. Fabrikant 1 Department of Geography, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland {annina.bruegger; kai-florian.richter; sara.fabrikant}@geo.uzh.ch Abstract We gladly use automated technology (e.g., smart devices) to extend our hard working minds. But what if such technology turns into mind crutches we cannot do without? Understanding how varying levels of automation in mobile maps might impact navigation performance and spatial knowledge acquisition will provide important insights for the ongoing debate on the potentially detrimental effects of using navigation systems on human spatial cognition. We need to identify the right balance between system automation (support) and user autonomy (self-reliance). Preliminary results of a pilot study performed within a novel empirical framework indicate that it is possible to increase user autonomy and spatial knowledge acquisition without negatively impacting navigation performance and usefulness of the system. 1. Introduction Various research fields have investigated how system automation might influence human knowledge and abilities. It is generally agreed that computers often make decisions originally performed by humans in a (more) efficient way. This has positive and negative impacts on humans as, for example, automation can reduce their physical and cognitive effort (Sheridan 2002). This also holds for navigational tasks, where acquiring spatial knowledge is crucial to orient and move in space without getting lost. Recent developments in self-driving vehicles highlight the need for better understanding human behavior, especially when humans have to take over from automated systems during system failure (Merat et al. 2014). The ideal human- system interaction would be to use the best of both human and technology (Sheridan 2002), which we aim for in our research. Specifically, how do we balance the advantages of system automation and the need for human autonomy to maximize both navigation efficiency and knowledge acquisition? 2. Balancing Assistance and Engagement Research investigating mobile navigation aids identified negative impacts on spatial knowledge acquisition, despite being very effective for efficient navigation (e.g. Willis et al. 2009). The consequences of automated guidance seem to be a disengagement of navigators’ attention from their surroundings (Gardony et al. 2013), and split attention between mobile device and the traversed environment (Willis et al. 2009). However, mobile navigation devices should enable pro-active engagement with the environment, which will lead to better spatial knowledge acquisition (Chung et al. 2016; Parush et al. 2007), as systems might break down, or users might lose the device and suddenly depend on their own abilities (Hirtle and Raubal 2013). The means to design such systems are yet unclear. Systems would need to provide efficient wayfinding support (sufficient system