Intro | Gesture | Navigation | Scaling | Visualizing 3D and Transformations | Category Adjustment Model
Spatial cognition occurs at many different scales. When humans interact with the world on the order of buildings, cities, or states (for example, locating a classroom, finding a friend’s house, or planning a road trip) the process is typically referred to as navigation. The projects in this section address questions about the types of things people use to navigate successfully, what makes some people better navigators than others, and the sorts of representation that might underlie navigation processes. We address these questions both to consider the cognitive processes that are important for navigation, and to identify tools or techniques to improve spatial cognition.
- Individual Differences in Using Slope for Navigation
- Slope as an Orienting Cue
- A Virtual Environment Assessment of Navigation Ability
- Using Analogy to Teach Topographic Maps
Individual Differences in Using Slope for Navigation (Steve Weisberg)
In previous navigation research, the ability to navigate has been attributed to three types of orientation cues: local landmarks, geometric peculiarities, and geocentric direction information. The gap in this paradigm is the dearth of research on the influence of slope on navigation. The only existing study by Restat et al. (2004) showed that participants were more accurate in pointing judgments and map drawing when they explored a sloped virtual environment rather than a flat one. We are interested to see if these effects are mediated by individual differences in navigation ability. Are people who are better navigators better at using slope as a navigational cue? To this end, we created sloped and non-sloped virtual environments for participants to explore, with the goal of learning the locations of various buildings. Our results so far have shown that many factors have played a role in the effectiveness of using slope as a navigation cue (e.g., self-reported navigation ability, gender). Now, we are looking to expand upon previous work by looking at other spatial factors that relate to navigation, as well as replicate previous findings. Future research will contrast slope with other types of cues that have been shown to be useful in navigation, and will look at the difference between slope as a directional cue (i.e., the building was uphill and to my left), versus slope as a local cue (i.e., the building was next to the big hill).
Figure 1. Sample screenshot of the sloped virtual environment.
A growing body of evidence has shown that non-human animals (rats: Miniaci, Scotto, & Bures, 1999; Moghaddam, Kaminsky, Zahalka, & Bures, 1996; crabs: Gherardi, Nocchi, & Vannini, 1988; pigeons: Nardi & Bingman, in press) and humans (Restat, Steck, Mochnatzki, & Mallot, 2004) can use terrain slope as an allocentric orienting cue for goal navigation. The salience of slope information derives from potentially redundant, multimodal sensory activations associated with the slope, such as visual, kinesthetic and vestibular stimuli.
So far, only one study was carried on the influence of terrain slope on human place-learning. Using a virtual reality environment it was found that, when a “virtual town” was presented on a slope, accuracy in navigation and in pointing to distant landmarks was massively improved with respect to a condition in which the virtual town was on a flat surface (Restat, Steck, Mochnatzki, & Mallot, 2004).
We aim to investigate the nature of the spatial representation that humans extract from terrain slope, using a real-world experimental environment that can be tilted. Among the possible studies that can be applied to this experimental apparatus are: slant perception, associative learning related to the interaction between slope and other spatial cues, and cue integration of spatial gradients such as slope. As a first step in this line of research, we are investigating to what degree humans can use a slope to encode a goal location.
A Virtual Environment Assessment of Navigation Ability (Steve Weisberg)
Research has shown that some people are much better navigators than others. In general, though, measuring navigation ability efficiently and accurately relies upon subjective self-reports, not objective, behavioral evidence. We created a virtual environment assessment of navigation ability that can be easily administered on a desktop computer to test whether an individual’s self-reported navigation ability, as well as other related spatial skills, reflected their actual ability to remember the layout of buildings in a large-scale virtual space. While there are certainly differences between finding one’s way in a virtual environment compared to the real world, the benefits of using a virtual environment for this project include the ease and efficiency of data collection; the inclusion of a large number of participants; and our ability to manipulate parameters of the environment to restrict or augment navigation cues. Our goals in creating this assessment tool are to understand the factors, both internal and external, that relate most directly to the ability to navigation ability, as well as understand whether navigation ability has direct relevance to skills that underlie success in the STEM disciplines. We also plan to launch the study online in the coming months, to gather data from a variety of populations.
Figure 1: Screenshots of buildings and routes along the virtual environment. Participants interacted with the virtual environment via arrows on a keyboard and a mouse, while viewing the environment on a standard desktop monitor.
For more information on this project see SILC June 2010 SHOWCASE "From the real to the virtual-world: Individual differences in navigation"
Using Analogy to Teach Topographic Maps (Steve Weisberg)
Contour maps represent three-dimensional (3D) terrain data about elevation on a two-dimensional (2D) display using contour lines (See figure 1 for an example). Despite being widely used in the geosciences, contour maps are incredibly difficult to master. Previous SILC research has demonstrated the viability of stereoscopic, or 3D, visualization as a strategy for improving contour map interpretation (Rapp et al., 2007), but only for maps where such 3D cues are available. This research takes a novel approach to teaching contour maps, focusing on the learning principles of analogy and progressive alignment to decompose the spatial visualization process, supporting student understanding of the relationship between the 3D environment, and the 2D contour lines. Because students often misinterpret rules about contour maps like “contour lines that are close together indicate a steep slope” (Clark et al., 2008), we are designing an experiment to test the applicability of analogy to promote student understanding of all contour maps.
Two training conditions will apply high and low alignable differences, as described by Genter’s structure-mapping theory (Gentner & Markman, 1983), to contour maps and photographs of the terrains to which they refer (Fig. 2). We predict that when the analogy between the map and the environment is alignable, students will learn principles of contour maps more easily, and performance will improve. If successful, this study will provide further demonstration of the robustness of these learning techniques, create an intervention that helps all students learn contour maps, and offer insight into the perception of 3D terrain as a 2D projection.
Sample topographic map with various features labeled.
Islands (left) with their corresponding contour map representations (right). The difference of the peak (top) from the plateau (bottom) is an alignable difference that could help students learn rules and features of topographic maps.