© The Eurographics Association 2007. IPT-EGVE Symposium (2007) Short Papers B. Fröhlich, R. Blach, and R. van Liere (Editors) A Quantitative Assessment of the Impact on Spatial Understanding of Exploring a Complex Immersive Virtual Environment using Augmented Real Walking versus Flying Victoria Interrante †1 , Eleanor O’Rourke 2 , Leanne Gray 3 , Lee Anderson 4 and Brian Ries 1 1 Department of Computer Science, University of Minnesota 2 Department of Computer Science, Colby College 3 Department of Computer Science, Kansas State University 4 Department of Architecture, University of Minnesota Abstract When an immersive virtual environment spans an area that is larger than the available physical space for real walking, one may use an ‘augmented walking’ method such as Seven League Boots to enable participants to explore the space while gaining proprioceptive feedback that is similar to what they would experience with normal walking. In this paper, we present the results of a preliminary experiment in which we seek to quantitatively assess the extent to which participants are able to make more accurate spatial judgments about the locations of previously-seen targets in a complicated virtual city environment, experienced using a head-mounted display, after traveling to them using augmented real walking (‘boots’) versus virtual walking enabled by a button press on a hand-held wand. In a series of trials, we ask participants to follow paths of increasing complexity from a home base to different hidden targets in the environment and back. At each endpoint, with the path markings turned off, we ask participants to point, through the intervening alleyway walls, to the location they believe they started from. Participants are able to make real turns with their bodies in both locomotion conditions, however they are able to make real forward movement only under the augmented walking condition. Each participant completes eight trials under each locomotion condition, with the target locations and the order of experiencing each method counterbalanced between participants. In data collected from six participants so far, we are finding that the median angle error is significantly greater, overall, in the wand locomotion condition than in the ‘boots’ locomotion condition, and that the errors tend to increase, overall, as the path complexity increases (from two segments to four segments) in the wand locomotion condition but not in the ‘boots’ locomotion condition. Categories and Subject Descriptors (according to ACM CCS): I.3.6 [Computer Graphics]: Methodology and Techniques 1. Introduction and Previous Work Previous research has shown that participants tend to achieve a greater sense of presence in an HMD-based immersive virtual environment when they are enabled to travel through that environment using actual real walking than when they must make walking motions in place (indirect walking) or resort to pressing a button on a wand (virtual walking) [UAWB*99]. However, when the immersive virtual environment spans an area that is larger than the physically available navigable tracked space, using real walking becomes problematic. Historically, the most common solution in those cases has been to either use an indirect technique for locomotion control, such as walking-in-place [TDS99] or leaning [LAKZ01], or to use a physical device such as an omni- directional treadmill [e.g. DCC97]. Recently, several alternative approaches that seek to more faithfully preserve the physical sensations associated with real walking have been proposed. The first is re-directed walking [RKW01], in which participants are surreptitiously induced to walk in a curving direction while believing that they are walking in a straight line. This method can be used to enable people to feel as if they are walking a long distance along a straight path, when they are actually just walking in a large circle. With the addition of large rotational distortion during large turns, this method can be successfully used to enable people to feel as if they are naturally walking between points in a space that is much more expansive than their actual traversed space. The second approach is augmented walking with seven league boots [IRA07], in which participants experience accelerated movement through the virtual environment in the primary direction of their actual travel. In a prior preference study, we found that participants rated their experience of exploring a virtual hallway using seven