SOLVING GEOSPATIAL TASKS IN 2D AND 3D VISUALIZATIONS OF MAPS - Information Visualisation: Techniques, Usability and Evaluation - page 227

C HAPTER T WELVE
S OLVING G EOSPATIAL T ASKS IN 2D AND 3D
V ISUALIZATIONS OF M APS
S TEFAN S EIPEL
Abstract
3D visualizations have previously been shown to be useful for solving a
number of tasks in various applications. In geovisualization, traditional 2D
maps provide fundamental spatial reference when presenting information.
This chapter presents a series of experiments that investigate different 3D
map representations for solving some generic spatial tasks. First, different
concepts of 3D visualizations are defined to establish the terminology used
in the subsequent descriptions of experiments. Two experiments are then
presented that investigate how users solve fundamental spatial assessment
tasks when 3D map representations are used to provide spatial context.
The main finding of the first experiment showed that comparative distance
assessment is not affected negatively by perspective distortion if maps are
slanted in 3D and if the slant angle is less than 65 degrees with regard to
the vertical plane. The results also confirmed a correlation between
accuracy and the similarity of distances to be assessed. The second
experiment investigated a more complex, combinatorial task that required
users to compare quantitative information on the map in relation to spatial
distance. A conventional 2D graphical mapping was compared with a 3D
graphical mapping in a slanted map. The results of this experiment did not
show any significant differences between the two methods for the entire
test population, although 83% of the subjects stated that they felt more
confident with the 3D visualization. Participants in the study who rated
their own spatial vision capabilities as high, performed significantly better
than those who rated them low. Also, frequent game players did not show
different results in terms of task performance as compared to non-frequent
game players.